Add model

config.json

@@ -0,0 +1,41 @@
+{
+  "architectures": [
+    "Gemma3ForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "attn_logit_softcapping": null,
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "eos_token_id": [
+    1,
+    106
+  ],
+  "final_logit_softcapping": null,
+  "head_dim": 128,
+  "hidden_activation": "gelu_pytorch_tanh",
+  "hidden_size": 5376,
+  "initializer_range": 0.02,
+  "intermediate_size": 21504,
+  "max_position_embeddings": 131072,
+  "mm_tokens_per_image": 256,
+  "model_type": "gemma3_text",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 62,
+  "num_key_value_heads": 16,
+  "pad_token_id": 0,
+  "query_pre_attn_scalar": 168,
+  "rms_norm_eps": 1e-06,
+  "rope_local_base_freq": 10000.0,
+  "rope_scaling": {
+    "factor": 8.0,
+    "rope_type": "linear"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 1024,
+  "sliding_window_pattern": 6,
+  "torch_dtype": "float32",
+  "transformers_version": "4.51.2",
+  "use_cache": false,
+  "vocab_size": 262208
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "bos_token_id": 2,
+  "cache_implementation": "hybrid",
+  "do_sample": true,
+  "eos_token_id": [
+    1,
+    106
+  ],
+  "pad_token_id": 0,
+  "top_k": 64,
+  "top_p": 0.95,
+  "transformers_version": "4.51.2"
+}

model-00001-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d66f397442bde61ca9f1be201d02d02b694445d2fdd4af45287c8c18519e83b0
+size 5638520968

model-00002-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8a9e1a05f9a64f88dcd3cd19b731c5e5e491be10c073292129f7dd4a06be9745
+size 4954787224

model-00003-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9afdc4c4f62f43864d007df69ec6d0ee2dd5db7e0a377746fb5d25a9d96a9c4c
+size 4954787224

model-00004-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faff429a1608ed1a573b80f4375f3e4bb43ab41882200b4140c38abee1fa1fce
+size 4954787224

model-00005-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89d7dcbd954e1c539cec8d3674375e225a1d3ee4ee25e35725b6a56bc593ef82
+size 4954787248

model-00006-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5b5ac6a8909004d10ce7e2047353f08e4f90fb58fe4ced26306f5b46fd50081c
+size 4954787264

model-00007-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b593b3e2b506a6bfde84a7d3d717399ca7f1fd05ff1a65a25403b364d18a6965
+size 4954787264

model-00008-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6842b19f56c679679c5f8a6b04185c5aa0ef568875d54c2958c29fe54045817f
+size 4954787264

model-00009-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d806e25d442326f58dbd36ddebb8e5935452faf7cd18ac820d5d5ba59b94406a
+size 4954787264

model-00010-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0617ec36e050fc179b3a664712faa2eeedb6507b70d01314916d525939cdd044
+size 4954787264

model-00011-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fadd3bc67c163753053ab20d332b01a4a938a5326b2d8f4d9a90c86915348c70
+size 4954787264

model-00012-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:961692cd058aa3327b1231aeb2c8b94dd60b95d37a139b49a17788d6407d9555
+size 4954787264

model-00013-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5fe6af2e34e2565cdde28c94a56b5c6d57b7140053e946e353f4b75e8626a3fe
+size 4954787264

model-00014-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3527bad5e35bcf0ab738dfe2f646c63a7268b0039a32e72aa5f2669441edec3f
+size 4954787264

model-00015-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ca5702ac9c1dd7813b3334fe13c4da0c77bedf0349fd6425ac0d9c5b9375fef
+size 4954787264

model-00016-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13654758718093fa41d428e7ec9c2bf877a1b330ff70daf16817ea2224ca5ed7
+size 4954787264

model-00017-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:49f0366db9dced195d21952688889a5c9176f8c323c0165032e7979c366dd628
+size 4954787264

model-00018-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c7f497666e03c0108d5cff1ea40f1ff8028801dc659008f5e07771b54555288
+size 4954787264

model-00019-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2bc50d37d05fc010cfe998a10dac6efe1733c29928d1001828e6ceb234a3dcb
+size 4954787264

model-00020-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32644bdd94eaf5bd9507262be8a57d2aaf287b3234930aa4106daf1c7e5c2679
+size 4954787264

model-00021-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ab8284f6dea57fc48a1f509ddc55f176c8fa019d9a7c3dc8a009135edd441e27
+size 4954787264

model-00022-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7be1032bcd11112833cbd84c2ad48b90e292630a27022f124d23a05d2580230b
+size 5638520960

model-00023-of-00023.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:df49a97045c52fc3518e77d0529c6f97a23921848e5a652d37e5e08951b45bb1
+size 3303213112

model.safetensors.index.json

@@ -0,0 +1,816 @@
+{
+  "metadata": {
+    "total_size": 113675906048
+  },
+  "weight_map": {
+    "lm_head.weight": "model-00022-of-00023.safetensors",
+    "model.embed_tokens.weight": "model-00001-of-00023.safetensors",
+    "model.layers.0.input_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.post_feedforward_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.pre_feedforward_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.self_attn.k_norm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.self_attn.q_norm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.input_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.post_feedforward_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.pre_feedforward_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.self_attn.k_norm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.self_attn.q_norm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.10.input_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.post_feedforward_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.pre_feedforward_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.self_attn.k_norm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.self_attn.q_norm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.input_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.post_feedforward_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.pre_feedforward_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.self_attn.k_norm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.self_attn.q_norm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.12.input_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.post_feedforward_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.pre_feedforward_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.self_attn.k_norm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.self_attn.q_norm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.input_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.post_feedforward_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.pre_feedforward_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.self_attn.k_norm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.self_attn.q_norm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.input_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.post_feedforward_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.pre_feedforward_layernorm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.self_attn.k_norm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.self_attn.q_norm.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "model-00006-of-00023.safetensors",
+    "model.layers.15.input_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.post_feedforward_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.pre_feedforward_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.self_attn.k_norm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.self_attn.q_norm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.input_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.post_feedforward_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.pre_feedforward_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.self_attn.k_norm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.self_attn.q_norm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.input_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.post_feedforward_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.pre_feedforward_layernorm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.self_attn.k_norm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.self_attn.q_norm.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "model-00007-of-00023.safetensors",
+    "model.layers.18.input_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.post_feedforward_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.pre_feedforward_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.self_attn.k_norm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.self_attn.q_norm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.input_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.post_feedforward_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.pre_feedforward_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.self_attn.k_norm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.self_attn.q_norm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.2.input_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.post_feedforward_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.pre_feedforward_layernorm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.self_attn.k_norm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.self_attn.q_norm.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "model-00002-of-00023.safetensors",
+    "model.layers.20.input_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.post_feedforward_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.pre_feedforward_layernorm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.self_attn.k_norm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.self_attn.q_norm.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "model-00008-of-00023.safetensors",
+    "model.layers.21.input_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.post_feedforward_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.pre_feedforward_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.self_attn.k_norm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.self_attn.q_norm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.input_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.post_feedforward_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.pre_feedforward_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.self_attn.k_norm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.self_attn.q_norm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.input_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.post_feedforward_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.pre_feedforward_layernorm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.self_attn.k_norm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.self_attn.q_norm.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "model-00009-of-00023.safetensors",
+    "model.layers.24.input_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.post_feedforward_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.pre_feedforward_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.self_attn.k_norm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.self_attn.q_norm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.input_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.post_feedforward_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.pre_feedforward_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.self_attn.k_norm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.self_attn.q_norm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.input_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.post_feedforward_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.pre_feedforward_layernorm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.self_attn.k_norm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.self_attn.q_norm.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "model-00010-of-00023.safetensors",
+    "model.layers.27.input_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.post_feedforward_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.pre_feedforward_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.self_attn.k_norm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.self_attn.q_norm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.input_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.post_feedforward_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.pre_feedforward_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.self_attn.k_norm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.self_attn.q_norm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.input_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.post_feedforward_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.pre_feedforward_layernorm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.self_attn.k_norm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.self_attn.q_norm.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "model-00011-of-00023.safetensors",
+    "model.layers.3.input_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.post_feedforward_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.pre_feedforward_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.self_attn.k_norm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.self_attn.q_norm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.30.input_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.post_feedforward_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.pre_feedforward_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.self_attn.k_norm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.self_attn.q_norm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.input_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.post_feedforward_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.pre_feedforward_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.self_attn.k_norm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.self_attn.q_norm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.input_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.mlp.down_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.mlp.gate_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.mlp.up_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.post_attention_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.post_feedforward_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.pre_feedforward_layernorm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.self_attn.k_norm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.self_attn.k_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.self_attn.o_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.self_attn.q_norm.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.self_attn.q_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.32.self_attn.v_proj.weight": "model-00012-of-00023.safetensors",
+    "model.layers.33.input_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.mlp.down_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.mlp.gate_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.mlp.up_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.post_attention_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.post_feedforward_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.pre_feedforward_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.self_attn.k_norm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.self_attn.k_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.self_attn.o_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.self_attn.q_norm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.self_attn.q_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.33.self_attn.v_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.input_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.mlp.down_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.mlp.gate_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.mlp.up_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.post_attention_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.post_feedforward_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.pre_feedforward_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.self_attn.k_norm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.self_attn.k_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.self_attn.o_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.self_attn.q_norm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.self_attn.q_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.34.self_attn.v_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.input_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.mlp.down_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.mlp.gate_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.mlp.up_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.post_attention_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.post_feedforward_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.pre_feedforward_layernorm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.self_attn.k_norm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.self_attn.k_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.self_attn.o_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.self_attn.q_norm.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.self_attn.q_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.35.self_attn.v_proj.weight": "model-00013-of-00023.safetensors",
+    "model.layers.36.input_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.mlp.down_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.mlp.gate_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.mlp.up_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.post_attention_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.post_feedforward_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.pre_feedforward_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.self_attn.k_norm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.self_attn.k_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.self_attn.o_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.self_attn.q_norm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.self_attn.q_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.36.self_attn.v_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.input_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.mlp.down_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.mlp.gate_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.mlp.up_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.post_attention_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.post_feedforward_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.pre_feedforward_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.self_attn.k_norm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.self_attn.k_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.self_attn.o_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.self_attn.q_norm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.self_attn.q_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.37.self_attn.v_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.input_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.mlp.down_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.mlp.gate_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.mlp.up_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.post_attention_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.post_feedforward_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.pre_feedforward_layernorm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.self_attn.k_norm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.self_attn.k_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.self_attn.o_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.self_attn.q_norm.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.self_attn.q_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.38.self_attn.v_proj.weight": "model-00014-of-00023.safetensors",
+    "model.layers.39.input_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.mlp.down_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.mlp.gate_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.mlp.up_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.post_attention_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.post_feedforward_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.pre_feedforward_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.self_attn.k_norm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.self_attn.k_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.self_attn.o_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.self_attn.q_norm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.self_attn.q_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.39.self_attn.v_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.4.input_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.post_feedforward_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.pre_feedforward_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.self_attn.k_norm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.self_attn.q_norm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.40.input_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.mlp.down_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.mlp.gate_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.mlp.up_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.post_attention_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.post_feedforward_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.pre_feedforward_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.self_attn.k_norm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.self_attn.k_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.self_attn.o_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.self_attn.q_norm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.self_attn.q_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.40.self_attn.v_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.input_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.mlp.down_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.mlp.gate_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.mlp.up_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.post_attention_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.post_feedforward_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.pre_feedforward_layernorm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.self_attn.k_norm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.self_attn.k_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.self_attn.o_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.self_attn.q_norm.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.self_attn.q_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.41.self_attn.v_proj.weight": "model-00015-of-00023.safetensors",
+    "model.layers.42.input_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.mlp.down_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.mlp.gate_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.mlp.up_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.post_attention_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.post_feedforward_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.pre_feedforward_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.self_attn.k_norm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.self_attn.k_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.self_attn.o_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.self_attn.q_norm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.self_attn.q_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.42.self_attn.v_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.input_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.mlp.down_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.mlp.gate_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.mlp.up_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.post_attention_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.post_feedforward_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.pre_feedforward_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.self_attn.k_norm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.self_attn.k_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.self_attn.o_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.self_attn.q_norm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.self_attn.q_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.43.self_attn.v_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.input_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.mlp.down_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.mlp.gate_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.mlp.up_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.post_attention_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.post_feedforward_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.pre_feedforward_layernorm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.self_attn.k_norm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.self_attn.k_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.self_attn.o_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.self_attn.q_norm.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.self_attn.q_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.44.self_attn.v_proj.weight": "model-00016-of-00023.safetensors",
+    "model.layers.45.input_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.mlp.down_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.mlp.gate_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.mlp.up_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.post_attention_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.post_feedforward_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.pre_feedforward_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.self_attn.k_norm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.self_attn.k_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.self_attn.o_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.self_attn.q_norm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.self_attn.q_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.45.self_attn.v_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.input_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.mlp.down_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.mlp.gate_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.mlp.up_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.post_attention_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.post_feedforward_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.pre_feedforward_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.self_attn.k_norm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.self_attn.k_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.self_attn.o_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.self_attn.q_norm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.self_attn.q_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.46.self_attn.v_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.input_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.mlp.down_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.mlp.gate_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.mlp.up_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.post_attention_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.post_feedforward_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.pre_feedforward_layernorm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.self_attn.k_norm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.self_attn.k_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.self_attn.o_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.self_attn.q_norm.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.self_attn.q_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.47.self_attn.v_proj.weight": "model-00017-of-00023.safetensors",
+    "model.layers.48.input_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.mlp.down_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.mlp.gate_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.mlp.up_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.post_attention_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.post_feedforward_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.pre_feedforward_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.self_attn.k_norm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.self_attn.k_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.self_attn.o_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.self_attn.q_norm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.self_attn.q_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.48.self_attn.v_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.input_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.mlp.down_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.mlp.gate_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.mlp.up_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.post_attention_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.post_feedforward_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.pre_feedforward_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.self_attn.k_norm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.self_attn.k_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.self_attn.o_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.self_attn.q_norm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.self_attn.q_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.49.self_attn.v_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.5.input_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.post_feedforward_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.pre_feedforward_layernorm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.self_attn.k_norm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.self_attn.q_norm.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "model-00003-of-00023.safetensors",
+    "model.layers.50.input_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.mlp.down_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.mlp.gate_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.mlp.up_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.post_attention_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.post_feedforward_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.pre_feedforward_layernorm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.self_attn.k_norm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.self_attn.k_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.self_attn.o_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.self_attn.q_norm.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.self_attn.q_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.50.self_attn.v_proj.weight": "model-00018-of-00023.safetensors",
+    "model.layers.51.input_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.mlp.down_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.mlp.gate_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.mlp.up_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.post_attention_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.post_feedforward_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.pre_feedforward_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.self_attn.k_norm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.self_attn.k_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.self_attn.o_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.self_attn.q_norm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.self_attn.q_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.51.self_attn.v_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.input_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.mlp.down_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.mlp.gate_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.mlp.up_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.post_attention_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.post_feedforward_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.pre_feedforward_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.self_attn.k_norm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.self_attn.k_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.self_attn.o_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.self_attn.q_norm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.self_attn.q_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.52.self_attn.v_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.input_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.mlp.down_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.mlp.gate_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.mlp.up_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.post_attention_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.post_feedforward_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.pre_feedforward_layernorm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.self_attn.k_norm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.self_attn.k_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.self_attn.o_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.self_attn.q_norm.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.self_attn.q_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.53.self_attn.v_proj.weight": "model-00019-of-00023.safetensors",
+    "model.layers.54.input_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.mlp.down_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.mlp.gate_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.mlp.up_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.post_attention_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.post_feedforward_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.pre_feedforward_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.self_attn.k_norm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.self_attn.k_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.self_attn.o_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.self_attn.q_norm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.self_attn.q_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.54.self_attn.v_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.input_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.mlp.down_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.mlp.gate_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.mlp.up_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.post_attention_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.post_feedforward_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.pre_feedforward_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.self_attn.k_norm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.self_attn.k_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.self_attn.o_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.self_attn.q_norm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.self_attn.q_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.55.self_attn.v_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.input_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.mlp.down_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.mlp.gate_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.mlp.up_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.post_attention_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.post_feedforward_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.pre_feedforward_layernorm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.self_attn.k_norm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.self_attn.k_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.self_attn.o_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.self_attn.q_norm.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.self_attn.q_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.56.self_attn.v_proj.weight": "model-00020-of-00023.safetensors",
+    "model.layers.57.input_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.mlp.down_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.mlp.gate_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.mlp.up_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.post_attention_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.post_feedforward_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.pre_feedforward_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.self_attn.k_norm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.self_attn.k_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.self_attn.o_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.self_attn.q_norm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.self_attn.q_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.57.self_attn.v_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.input_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.mlp.down_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.mlp.gate_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.mlp.up_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.post_attention_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.post_feedforward_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.pre_feedforward_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.self_attn.k_norm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.self_attn.k_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.self_attn.o_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.self_attn.q_norm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.self_attn.q_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.58.self_attn.v_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.input_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.mlp.down_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.mlp.gate_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.mlp.up_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.post_attention_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.post_feedforward_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.pre_feedforward_layernorm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.self_attn.k_norm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.self_attn.k_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.self_attn.o_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.self_attn.q_norm.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.self_attn.q_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.59.self_attn.v_proj.weight": "model-00021-of-00023.safetensors",
+    "model.layers.6.input_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.post_feedforward_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.pre_feedforward_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.self_attn.k_norm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.self_attn.q_norm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.60.input_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.mlp.down_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.mlp.gate_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.mlp.up_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.post_attention_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.post_feedforward_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.pre_feedforward_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.self_attn.k_norm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.self_attn.k_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.self_attn.o_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.self_attn.q_norm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.self_attn.q_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.60.self_attn.v_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.input_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.mlp.down_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.mlp.gate_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.mlp.up_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.post_attention_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.post_feedforward_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.pre_feedforward_layernorm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.self_attn.k_norm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.self_attn.k_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.self_attn.o_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.self_attn.q_norm.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.self_attn.q_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.61.self_attn.v_proj.weight": "model-00023-of-00023.safetensors",
+    "model.layers.7.input_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.post_feedforward_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.pre_feedforward_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.self_attn.k_norm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.self_attn.q_norm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.input_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.post_feedforward_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.pre_feedforward_layernorm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.self_attn.k_norm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.self_attn.q_norm.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "model-00004-of-00023.safetensors",
+    "model.layers.9.input_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.post_feedforward_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.pre_feedforward_layernorm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.self_attn.k_norm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.self_attn.q_norm.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "model-00005-of-00023.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "model-00005-of-00023.safetensors",
+    "model.norm.weight": "model-00023-of-00023.safetensors"
+  }
+}

special_tokens_map.json

@@ -0,0 +1,33 @@
+{
+  "boi_token": "<start_of_image>",
+  "bos_token": {
+    "content": "<bos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eoi_token": "<end_of_image>",
+  "eos_token": {
+    "content": "<eos>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "image_token": "<image_soft_token>",
+  "pad_token": {
+    "content": "<pad>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e316dda1f34f410e1a27d086d70cbb20164153422fcae753e180a47b083c99b3
+size 18452294

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1299c11d7cf632ef3b4e11937501358ada021bbdf7c47638d13c0ee982f2e79c
+size 4689074

trainer_state.json

@@ -0,0 +1,2876 @@
+{
+  "best_global_step": null,
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 0,
+  "global_step": 406,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.0024630541871921183,
+      "grad_norm": 11.540680885314941,
+      "learning_rate": 1e-05,
+      "loss": 0.9851,
+      "step": 1
+    },
+    {
+      "epoch": 0.0049261083743842365,
+      "grad_norm": 4.776151657104492,
+      "learning_rate": 9.999850312505222e-06,
+      "loss": 0.9615,
+      "step": 2
+    },
+    {
+      "epoch": 0.007389162561576354,
+      "grad_norm": 3.0982847213745117,
+      "learning_rate": 9.999401258983426e-06,
+      "loss": 0.9231,
+      "step": 3
+    },
+    {
+      "epoch": 0.009852216748768473,
+      "grad_norm": 2.3222382068634033,
+      "learning_rate": 9.998652866321688e-06,
+      "loss": 0.7308,
+      "step": 4
+    },
+    {
+      "epoch": 0.012315270935960592,
+      "grad_norm": 1.997889757156372,
+      "learning_rate": 9.997605179330018e-06,
+      "loss": 0.7855,
+      "step": 5
+    },
+    {
+      "epoch": 0.014778325123152709,
+      "grad_norm": 1.7656501531600952,
+      "learning_rate": 9.996258260738676e-06,
+      "loss": 0.8678,
+      "step": 6
+    },
+    {
+      "epoch": 0.017241379310344827,
+      "grad_norm": 1.3362767696380615,
+      "learning_rate": 9.994612191194407e-06,
+      "loss": 0.7193,
+      "step": 7
+    },
+    {
+      "epoch": 0.019704433497536946,
+      "grad_norm": 2.2548017501831055,
+      "learning_rate": 9.99266706925562e-06,
+      "loss": 0.6996,
+      "step": 8
+    },
+    {
+      "epoch": 0.022167487684729065,
+      "grad_norm": 1.2499150037765503,
+      "learning_rate": 9.990423011386489e-06,
+      "loss": 0.7632,
+      "step": 9
+    },
+    {
+      "epoch": 0.024630541871921183,
+      "grad_norm": 1.1288288831710815,
+      "learning_rate": 9.987880151949976e-06,
+      "loss": 0.7409,
+      "step": 10
+    },
+    {
+      "epoch": 0.027093596059113302,
+      "grad_norm": 1.0057581663131714,
+      "learning_rate": 9.98503864319978e-06,
+      "loss": 0.7303,
+      "step": 11
+    },
+    {
+      "epoch": 0.029556650246305417,
+      "grad_norm": 0.968294084072113,
+      "learning_rate": 9.981898655271237e-06,
+      "loss": 0.6851,
+      "step": 12
+    },
+    {
+      "epoch": 0.03201970443349754,
+      "grad_norm": 1.1969448328018188,
+      "learning_rate": 9.978460376171113e-06,
+      "loss": 0.7647,
+      "step": 13
+    },
+    {
+      "epoch": 0.034482758620689655,
+      "grad_norm": 0.983593225479126,
+      "learning_rate": 9.974724011766364e-06,
+      "loss": 0.7308,
+      "step": 14
+    },
+    {
+      "epoch": 0.03694581280788178,
+      "grad_norm": 1.9946510791778564,
+      "learning_rate": 9.970689785771798e-06,
+      "loss": 0.6388,
+      "step": 15
+    },
+    {
+      "epoch": 0.03940886699507389,
+      "grad_norm": 1.070317029953003,
+      "learning_rate": 9.966357939736692e-06,
+      "loss": 0.6595,
+      "step": 16
+    },
+    {
+      "epoch": 0.04187192118226601,
+      "grad_norm": 1.241529941558838,
+      "learning_rate": 9.961728733030318e-06,
+      "loss": 0.8895,
+      "step": 17
+    },
+    {
+      "epoch": 0.04433497536945813,
+      "grad_norm": 1.0109727382659912,
+      "learning_rate": 9.956802442826417e-06,
+      "loss": 0.7378,
+      "step": 18
+    },
+    {
+      "epoch": 0.046798029556650245,
+      "grad_norm": 1.079637050628662,
+      "learning_rate": 9.951579364086603e-06,
+      "loss": 0.7178,
+      "step": 19
+    },
+    {
+      "epoch": 0.04926108374384237,
+      "grad_norm": 1.2305184602737427,
+      "learning_rate": 9.946059809542706e-06,
+      "loss": 0.5985,
+      "step": 20
+    },
+    {
+      "epoch": 0.05172413793103448,
+      "grad_norm": 11.067344665527344,
+      "learning_rate": 9.940244109678043e-06,
+      "loss": 0.681,
+      "step": 21
+    },
+    {
+      "epoch": 0.054187192118226604,
+      "grad_norm": 1.4613230228424072,
+      "learning_rate": 9.934132612707631e-06,
+      "loss": 0.6258,
+      "step": 22
+    },
+    {
+      "epoch": 0.05665024630541872,
+      "grad_norm": 1.041548728942871,
+      "learning_rate": 9.927725684557339e-06,
+      "loss": 0.5983,
+      "step": 23
+    },
+    {
+      "epoch": 0.059113300492610835,
+      "grad_norm": 1.0944949388504028,
+      "learning_rate": 9.921023708841975e-06,
+      "loss": 0.6303,
+      "step": 24
+    },
+    {
+      "epoch": 0.06157635467980296,
+      "grad_norm": 0.964139461517334,
+      "learning_rate": 9.914027086842323e-06,
+      "loss": 0.6394,
+      "step": 25
+    },
+    {
+      "epoch": 0.06403940886699508,
+      "grad_norm": 1.1896731853485107,
+      "learning_rate": 9.90673623748111e-06,
+      "loss": 0.6009,
+      "step": 26
+    },
+    {
+      "epoch": 0.0665024630541872,
+      "grad_norm": 1.1914570331573486,
+      "learning_rate": 9.899151597297923e-06,
+      "loss": 0.6132,
+      "step": 27
+    },
+    {
+      "epoch": 0.06896551724137931,
+      "grad_norm": 0.910982608795166,
+      "learning_rate": 9.891273620423083e-06,
+      "loss": 0.6051,
+      "step": 28
+    },
+    {
+      "epoch": 0.07142857142857142,
+      "grad_norm": 0.9912497401237488,
+      "learning_rate": 9.883102778550434e-06,
+      "loss": 0.5512,
+      "step": 29
+    },
+    {
+      "epoch": 0.07389162561576355,
+      "grad_norm": 1.612174153327942,
+      "learning_rate": 9.874639560909118e-06,
+      "loss": 0.5836,
+      "step": 30
+    },
+    {
+      "epoch": 0.07635467980295567,
+      "grad_norm": 0.967313289642334,
+      "learning_rate": 9.865884474234275e-06,
+      "loss": 0.6504,
+      "step": 31
+    },
+    {
+      "epoch": 0.07881773399014778,
+      "grad_norm": 4.314541339874268,
+      "learning_rate": 9.856838042736698e-06,
+      "loss": 0.654,
+      "step": 32
+    },
+    {
+      "epoch": 0.0812807881773399,
+      "grad_norm": 0.9903213977813721,
+      "learning_rate": 9.847500808071458e-06,
+      "loss": 0.5655,
+      "step": 33
+    },
+    {
+      "epoch": 0.08374384236453201,
+      "grad_norm": 0.9885880947113037,
+      "learning_rate": 9.837873329305458e-06,
+      "loss": 0.6559,
+      "step": 34
+    },
+    {
+      "epoch": 0.08620689655172414,
+      "grad_norm": 0.9196619987487793,
+      "learning_rate": 9.82795618288397e-06,
+      "loss": 0.5714,
+      "step": 35
+    },
+    {
+      "epoch": 0.08866995073891626,
+      "grad_norm": 0.9677406549453735,
+      "learning_rate": 9.817749962596115e-06,
+      "loss": 0.5687,
+      "step": 36
+    },
+    {
+      "epoch": 0.09113300492610837,
+      "grad_norm": 1.034245491027832,
+      "learning_rate": 9.807255279539313e-06,
+      "loss": 0.6626,
+      "step": 37
+    },
+    {
+      "epoch": 0.09359605911330049,
+      "grad_norm": 1.936618685722351,
+      "learning_rate": 9.796472762082687e-06,
+      "loss": 0.6928,
+      "step": 38
+    },
+    {
+      "epoch": 0.0960591133004926,
+      "grad_norm": 0.9693952202796936,
+      "learning_rate": 9.78540305582945e-06,
+      "loss": 0.5772,
+      "step": 39
+    },
+    {
+      "epoch": 0.09852216748768473,
+      "grad_norm": 1.231882929801941,
+      "learning_rate": 9.77404682357824e-06,
+      "loss": 0.6339,
+      "step": 40
+    },
+    {
+      "epoch": 0.10098522167487685,
+      "grad_norm": 0.9423834681510925,
+      "learning_rate": 9.762404745283439e-06,
+      "loss": 0.5313,
+      "step": 41
+    },
+    {
+      "epoch": 0.10344827586206896,
+      "grad_norm": 1.3421170711517334,
+      "learning_rate": 9.75047751801446e-06,
+      "loss": 0.6631,
+      "step": 42
+    },
+    {
+      "epoch": 0.10591133004926108,
+      "grad_norm": 1.0659502744674683,
+      "learning_rate": 9.738265855914014e-06,
+      "loss": 0.6041,
+      "step": 43
+    },
+    {
+      "epoch": 0.10837438423645321,
+      "grad_norm": 14.726941108703613,
+      "learning_rate": 9.725770490155338e-06,
+      "loss": 0.4993,
+      "step": 44
+    },
+    {
+      "epoch": 0.11083743842364532,
+      "grad_norm": 1.113370418548584,
+      "learning_rate": 9.712992168898436e-06,
+      "loss": 0.6032,
+      "step": 45
+    },
+    {
+      "epoch": 0.11330049261083744,
+      "grad_norm": 4.380392074584961,
+      "learning_rate": 9.699931657245264e-06,
+      "loss": 0.5649,
+      "step": 46
+    },
+    {
+      "epoch": 0.11576354679802955,
+      "grad_norm": 1.0713157653808594,
+      "learning_rate": 9.686589737193929e-06,
+      "loss": 0.5833,
+      "step": 47
+    },
+    {
+      "epoch": 0.11822660098522167,
+      "grad_norm": 0.79201340675354,
+      "learning_rate": 9.67296720759187e-06,
+      "loss": 0.6057,
+      "step": 48
+    },
+    {
+      "epoch": 0.1206896551724138,
+      "grad_norm": 0.8670496940612793,
+      "learning_rate": 9.659064884088017e-06,
+      "loss": 0.5203,
+      "step": 49
+    },
+    {
+      "epoch": 0.12315270935960591,
+      "grad_norm": 0.9468395709991455,
+      "learning_rate": 9.644883599083959e-06,
+      "loss": 0.6177,
+      "step": 50
+    },
+    {
+      "epoch": 0.12561576354679804,
+      "grad_norm": 1.1626007556915283,
+      "learning_rate": 9.630424201684105e-06,
+      "loss": 0.5259,
+      "step": 51
+    },
+    {
+      "epoch": 0.12807881773399016,
+      "grad_norm": 0.8885717391967773,
+      "learning_rate": 9.615687557644848e-06,
+      "loss": 0.5835,
+      "step": 52
+    },
+    {
+      "epoch": 0.13054187192118227,
+      "grad_norm": 0.9217110872268677,
+      "learning_rate": 9.600674549322716e-06,
+      "loss": 0.5965,
+      "step": 53
+    },
+    {
+      "epoch": 0.1330049261083744,
+      "grad_norm": 0.9233891367912292,
+      "learning_rate": 9.585386075621553e-06,
+      "loss": 0.5679,
+      "step": 54
+    },
+    {
+      "epoch": 0.1354679802955665,
+      "grad_norm": 1.285866141319275,
+      "learning_rate": 9.569823051938689e-06,
+      "loss": 0.5387,
+      "step": 55
+    },
+    {
+      "epoch": 0.13793103448275862,
+      "grad_norm": 0.8282471895217896,
+      "learning_rate": 9.553986410110135e-06,
+      "loss": 0.476,
+      "step": 56
+    },
+    {
+      "epoch": 0.14039408866995073,
+      "grad_norm": 0.9375513792037964,
+      "learning_rate": 9.537877098354787e-06,
+      "loss": 0.5456,
+      "step": 57
+    },
+    {
+      "epoch": 0.14285714285714285,
+      "grad_norm": 1.073931097984314,
+      "learning_rate": 9.521496081217652e-06,
+      "loss": 0.4895,
+      "step": 58
+    },
+    {
+      "epoch": 0.14532019704433496,
+      "grad_norm": 0.9736307263374329,
+      "learning_rate": 9.504844339512096e-06,
+      "loss": 0.6053,
+      "step": 59
+    },
+    {
+      "epoch": 0.1477832512315271,
+      "grad_norm": 1.0761698484420776,
+      "learning_rate": 9.487922870261123e-06,
+      "loss": 0.4654,
+      "step": 60
+    },
+    {
+      "epoch": 0.15024630541871922,
+      "grad_norm": 0.9952367544174194,
+      "learning_rate": 9.470732686637665e-06,
+      "loss": 0.5585,
+      "step": 61
+    },
+    {
+      "epoch": 0.15270935960591134,
+      "grad_norm": 0.8771824836730957,
+      "learning_rate": 9.453274817903932e-06,
+      "loss": 0.5827,
+      "step": 62
+    },
+    {
+      "epoch": 0.15517241379310345,
+      "grad_norm": 1.1052207946777344,
+      "learning_rate": 9.435550309349776e-06,
+      "loss": 0.5705,
+      "step": 63
+    },
+    {
+      "epoch": 0.15763546798029557,
+      "grad_norm": 0.8488280177116394,
+      "learning_rate": 9.417560222230115e-06,
+      "loss": 0.5058,
+      "step": 64
+    },
+    {
+      "epoch": 0.16009852216748768,
+      "grad_norm": 0.9430592656135559,
+      "learning_rate": 9.399305633701372e-06,
+      "loss": 0.506,
+      "step": 65
+    },
+    {
+      "epoch": 0.1625615763546798,
+      "grad_norm": 1.6385177373886108,
+      "learning_rate": 9.380787636757002e-06,
+      "loss": 0.6934,
+      "step": 66
+    },
+    {
+      "epoch": 0.16502463054187191,
+      "grad_norm": 0.9341111183166504,
+      "learning_rate": 9.36200734016203e-06,
+      "loss": 0.6388,
+      "step": 67
+    },
+    {
+      "epoch": 0.16748768472906403,
+      "grad_norm": 0.7698712944984436,
+      "learning_rate": 9.342965868386674e-06,
+      "loss": 0.4639,
+      "step": 68
+    },
+    {
+      "epoch": 0.16995073891625614,
+      "grad_norm": 0.8888192176818848,
+      "learning_rate": 9.32366436153902e-06,
+      "loss": 0.5409,
+      "step": 69
+    },
+    {
+      "epoch": 0.1724137931034483,
+      "grad_norm": 1.2890245914459229,
+      "learning_rate": 9.30410397529675e-06,
+      "loss": 0.6643,
+      "step": 70
+    },
+    {
+      "epoch": 0.1748768472906404,
+      "grad_norm": 0.8433436155319214,
+      "learning_rate": 9.284285880837947e-06,
+      "loss": 0.5345,
+      "step": 71
+    },
+    {
+      "epoch": 0.17733990147783252,
+      "grad_norm": 0.8337154388427734,
+      "learning_rate": 9.264211264770977e-06,
+      "loss": 0.4866,
+      "step": 72
+    },
+    {
+      "epoch": 0.17980295566502463,
+      "grad_norm": 0.8894972801208496,
+      "learning_rate": 9.243881329063436e-06,
+      "loss": 0.5478,
+      "step": 73
+    },
+    {
+      "epoch": 0.18226600985221675,
+      "grad_norm": 0.8818243741989136,
+      "learning_rate": 9.22329729097018e-06,
+      "loss": 0.5189,
+      "step": 74
+    },
+    {
+      "epoch": 0.18472906403940886,
+      "grad_norm": 1.0139381885528564,
+      "learning_rate": 9.202460382960449e-06,
+      "loss": 0.5413,
+      "step": 75
+    },
+    {
+      "epoch": 0.18719211822660098,
+      "grad_norm": 1.144716739654541,
+      "learning_rate": 9.181371852644063e-06,
+      "loss": 0.5689,
+      "step": 76
+    },
+    {
+      "epoch": 0.1896551724137931,
+      "grad_norm": 1.1229621171951294,
+      "learning_rate": 9.160032962696734e-06,
+      "loss": 0.5018,
+      "step": 77
+    },
+    {
+      "epoch": 0.1921182266009852,
+      "grad_norm": 1.20614755153656,
+      "learning_rate": 9.138444990784455e-06,
+      "loss": 0.5297,
+      "step": 78
+    },
+    {
+      "epoch": 0.19458128078817735,
+      "grad_norm": 1.7722558975219727,
+      "learning_rate": 9.116609229486992e-06,
+      "loss": 0.4937,
+      "step": 79
+    },
+    {
+      "epoch": 0.19704433497536947,
+      "grad_norm": 1.1039201021194458,
+      "learning_rate": 9.094526986220513e-06,
+      "loss": 0.5181,
+      "step": 80
+    },
+    {
+      "epoch": 0.19950738916256158,
+      "grad_norm": 1.0505284070968628,
+      "learning_rate": 9.072199583159285e-06,
+      "loss": 0.6021,
+      "step": 81
+    },
+    {
+      "epoch": 0.2019704433497537,
+      "grad_norm": 0.9036574959754944,
+      "learning_rate": 9.049628357156522e-06,
+      "loss": 0.5434,
+      "step": 82
+    },
+    {
+      "epoch": 0.2044334975369458,
+      "grad_norm": 1.5033870935440063,
+      "learning_rate": 9.026814659664331e-06,
+      "loss": 0.462,
+      "step": 83
+    },
+    {
+      "epoch": 0.20689655172413793,
+      "grad_norm": 0.9125608205795288,
+      "learning_rate": 9.003759856652803e-06,
+      "loss": 0.456,
+      "step": 84
+    },
+    {
+      "epoch": 0.20935960591133004,
+      "grad_norm": 3.8365941047668457,
+      "learning_rate": 8.98046532852822e-06,
+      "loss": 0.4463,
+      "step": 85
+    },
+    {
+      "epoch": 0.21182266009852216,
+      "grad_norm": 1.0002238750457764,
+      "learning_rate": 8.956932470050405e-06,
+      "loss": 0.5616,
+      "step": 86
+    },
+    {
+      "epoch": 0.21428571428571427,
+      "grad_norm": 0.807572603225708,
+      "learning_rate": 8.93316269024921e-06,
+      "loss": 0.4795,
+      "step": 87
+    },
+    {
+      "epoch": 0.21674876847290642,
+      "grad_norm": 0.7959713935852051,
+      "learning_rate": 8.90915741234015e-06,
+      "loss": 0.4098,
+      "step": 88
+    },
+    {
+      "epoch": 0.21921182266009853,
+      "grad_norm": 2.6254165172576904,
+      "learning_rate": 8.88491807363919e-06,
+      "loss": 0.6164,
+      "step": 89
+    },
+    {
+      "epoch": 0.22167487684729065,
+      "grad_norm": 0.8573901653289795,
+      "learning_rate": 8.860446125476688e-06,
+      "loss": 0.5426,
+      "step": 90
+    },
+    {
+      "epoch": 0.22413793103448276,
+      "grad_norm": 0.8792327046394348,
+      "learning_rate": 8.835743033110482e-06,
+      "loss": 0.5441,
+      "step": 91
+    },
+    {
+      "epoch": 0.22660098522167488,
+      "grad_norm": 1.1924870014190674,
+      "learning_rate": 8.810810275638183e-06,
+      "loss": 0.4542,
+      "step": 92
+    },
+    {
+      "epoch": 0.229064039408867,
+      "grad_norm": 0.7673595547676086,
+      "learning_rate": 8.78564934590859e-06,
+      "loss": 0.4864,
+      "step": 93
+    },
+    {
+      "epoch": 0.2315270935960591,
+      "grad_norm": 1.037055492401123,
+      "learning_rate": 8.760261750432312e-06,
+      "loss": 0.5286,
+      "step": 94
+    },
+    {
+      "epoch": 0.23399014778325122,
+      "grad_norm": 1.1282660961151123,
+      "learning_rate": 8.734649009291586e-06,
+      "loss": 0.5892,
+      "step": 95
+    },
+    {
+      "epoch": 0.23645320197044334,
+      "grad_norm": 0.9695258140563965,
+      "learning_rate": 8.708812656049227e-06,
+      "loss": 0.5851,
+      "step": 96
+    },
+    {
+      "epoch": 0.23891625615763548,
+      "grad_norm": 0.9574100971221924,
+      "learning_rate": 8.68275423765683e-06,
+      "loss": 0.5058,
+      "step": 97
+    },
+    {
+      "epoch": 0.2413793103448276,
+      "grad_norm": 0.8745037317276001,
+      "learning_rate": 8.656475314362149e-06,
+      "loss": 0.4157,
+      "step": 98
+    },
+    {
+      "epoch": 0.2438423645320197,
+      "grad_norm": 0.8950684070587158,
+      "learning_rate": 8.629977459615655e-06,
+      "loss": 0.4444,
+      "step": 99
+    },
+    {
+      "epoch": 0.24630541871921183,
+      "grad_norm": 1.0956140756607056,
+      "learning_rate": 8.603262259976348e-06,
+      "loss": 0.5102,
+      "step": 100
+    },
+    {
+      "epoch": 0.24876847290640394,
+      "grad_norm": 0.9333466291427612,
+      "learning_rate": 8.576331315016753e-06,
+      "loss": 0.5257,
+      "step": 101
+    },
+    {
+      "epoch": 0.2512315270935961,
+      "grad_norm": 0.9434043169021606,
+      "learning_rate": 8.549186237227138e-06,
+      "loss": 0.532,
+      "step": 102
+    },
+    {
+      "epoch": 0.2536945812807882,
+      "grad_norm": 1.0310450792312622,
+      "learning_rate": 8.521828651918983e-06,
+      "loss": 0.5588,
+      "step": 103
+    },
+    {
+      "epoch": 0.2561576354679803,
+      "grad_norm": 1.0325108766555786,
+      "learning_rate": 8.49426019712765e-06,
+      "loss": 0.4569,
+      "step": 104
+    },
+    {
+      "epoch": 0.25862068965517243,
+      "grad_norm": 1.030055284500122,
+      "learning_rate": 8.46648252351431e-06,
+      "loss": 0.5172,
+      "step": 105
+    },
+    {
+      "epoch": 0.26108374384236455,
+      "grad_norm": 0.8964868783950806,
+      "learning_rate": 8.438497294267117e-06,
+      "loss": 0.5143,
+      "step": 106
+    },
+    {
+      "epoch": 0.26354679802955666,
+      "grad_norm": 1.2119203805923462,
+      "learning_rate": 8.41030618500161e-06,
+      "loss": 0.4589,
+      "step": 107
+    },
+    {
+      "epoch": 0.2660098522167488,
+      "grad_norm": 0.880455732345581,
+      "learning_rate": 8.3819108836604e-06,
+      "loss": 0.4789,
+      "step": 108
+    },
+    {
+      "epoch": 0.2684729064039409,
+      "grad_norm": 1.026513695716858,
+      "learning_rate": 8.353313090412093e-06,
+      "loss": 0.4132,
+      "step": 109
+    },
+    {
+      "epoch": 0.270935960591133,
+      "grad_norm": 0.8435222506523132,
+      "learning_rate": 8.3245145175495e-06,
+      "loss": 0.4623,
+      "step": 110
+    },
+    {
+      "epoch": 0.2733990147783251,
+      "grad_norm": 0.8828105926513672,
+      "learning_rate": 8.295516889387115e-06,
+      "loss": 0.4267,
+      "step": 111
+    },
+    {
+      "epoch": 0.27586206896551724,
+      "grad_norm": 0.7920165061950684,
+      "learning_rate": 8.26632194215786e-06,
+      "loss": 0.4472,
+      "step": 112
+    },
+    {
+      "epoch": 0.27832512315270935,
+      "grad_norm": 1.1182401180267334,
+      "learning_rate": 8.23693142390914e-06,
+      "loss": 0.4812,
+      "step": 113
+    },
+    {
+      "epoch": 0.28078817733990147,
+      "grad_norm": 1.444700002670288,
+      "learning_rate": 8.207347094398173e-06,
+      "loss": 0.4815,
+      "step": 114
+    },
+    {
+      "epoch": 0.2832512315270936,
+      "grad_norm": 0.9819019436836243,
+      "learning_rate": 8.177570724986627e-06,
+      "loss": 0.47,
+      "step": 115
+    },
+    {
+      "epoch": 0.2857142857142857,
+      "grad_norm": 0.84306401014328,
+      "learning_rate": 8.14760409853456e-06,
+      "loss": 0.547,
+      "step": 116
+    },
+    {
+      "epoch": 0.2881773399014778,
+      "grad_norm": 0.9470686912536621,
+      "learning_rate": 8.117449009293668e-06,
+      "loss": 0.4874,
+      "step": 117
+    },
+    {
+      "epoch": 0.29064039408866993,
+      "grad_norm": 0.7772971987724304,
+      "learning_rate": 8.087107262799856e-06,
+      "loss": 0.5282,
+      "step": 118
+    },
+    {
+      "epoch": 0.29310344827586204,
+      "grad_norm": 0.9155469536781311,
+      "learning_rate": 8.05658067576513e-06,
+      "loss": 0.4475,
+      "step": 119
+    },
+    {
+      "epoch": 0.2955665024630542,
+      "grad_norm": 0.823267936706543,
+      "learning_rate": 8.025871075968828e-06,
+      "loss": 0.4273,
+      "step": 120
+    },
+    {
+      "epoch": 0.29802955665024633,
+      "grad_norm": 0.921981155872345,
+      "learning_rate": 7.99498030214817e-06,
+      "loss": 0.5247,
+      "step": 121
+    },
+    {
+      "epoch": 0.30049261083743845,
+      "grad_norm": 0.8674582839012146,
+      "learning_rate": 7.963910203888177e-06,
+      "loss": 0.4733,
+      "step": 122
+    },
+    {
+      "epoch": 0.30295566502463056,
+      "grad_norm": 0.895481526851654,
+      "learning_rate": 7.932662641510915e-06,
+      "loss": 0.5376,
+      "step": 123
+    },
+    {
+      "epoch": 0.3054187192118227,
+      "grad_norm": 0.7914291024208069,
+      "learning_rate": 7.90123948596412e-06,
+      "loss": 0.4572,
+      "step": 124
+    },
+    {
+      "epoch": 0.3078817733990148,
+      "grad_norm": 1.2109417915344238,
+      "learning_rate": 7.869642618709162e-06,
+      "loss": 0.4873,
+      "step": 125
+    },
+    {
+      "epoch": 0.3103448275862069,
+      "grad_norm": 0.9359944462776184,
+      "learning_rate": 7.8378739316084e-06,
+      "loss": 0.4277,
+      "step": 126
+    },
+    {
+      "epoch": 0.312807881773399,
+      "grad_norm": 0.8828278183937073,
+      "learning_rate": 7.805935326811913e-06,
+      "loss": 0.5501,
+      "step": 127
+    },
+    {
+      "epoch": 0.31527093596059114,
+      "grad_norm": 1.0584925413131714,
+      "learning_rate": 7.773828716643592e-06,
+      "loss": 0.4268,
+      "step": 128
+    },
+    {
+      "epoch": 0.31773399014778325,
+      "grad_norm": 1.5920400619506836,
+      "learning_rate": 7.741556023486655e-06,
+      "loss": 0.5066,
+      "step": 129
+    },
+    {
+      "epoch": 0.32019704433497537,
+      "grad_norm": 0.8200149536132812,
+      "learning_rate": 7.709119179668538e-06,
+      "loss": 0.419,
+      "step": 130
+    },
+    {
+      "epoch": 0.3226600985221675,
+      "grad_norm": 1.1859734058380127,
+      "learning_rate": 7.676520127345198e-06,
+      "loss": 0.496,
+      "step": 131
+    },
+    {
+      "epoch": 0.3251231527093596,
+      "grad_norm": 0.9637341499328613,
+      "learning_rate": 7.64376081838482e-06,
+      "loss": 0.5875,
+      "step": 132
+    },
+    {
+      "epoch": 0.3275862068965517,
+      "grad_norm": 0.9584794640541077,
+      "learning_rate": 7.610843214250964e-06,
+      "loss": 0.4876,
+      "step": 133
+    },
+    {
+      "epoch": 0.33004926108374383,
+      "grad_norm": 0.9975658059120178,
+      "learning_rate": 7.57776928588511e-06,
+      "loss": 0.4481,
+      "step": 134
+    },
+    {
+      "epoch": 0.33251231527093594,
+      "grad_norm": 1.1614975929260254,
+      "learning_rate": 7.5445410135886455e-06,
+      "loss": 0.5802,
+      "step": 135
+    },
+    {
+      "epoch": 0.33497536945812806,
+      "grad_norm": 2.236684799194336,
+      "learning_rate": 7.511160386904306e-06,
+      "loss": 0.4955,
+      "step": 136
+    },
+    {
+      "epoch": 0.3374384236453202,
+      "grad_norm": 0.8656754493713379,
+      "learning_rate": 7.477629404497048e-06,
+      "loss": 0.5312,
+      "step": 137
+    },
+    {
+      "epoch": 0.3399014778325123,
+      "grad_norm": 0.7941805124282837,
+      "learning_rate": 7.4439500740343685e-06,
+      "loss": 0.4797,
+      "step": 138
+    },
+    {
+      "epoch": 0.34236453201970446,
+      "grad_norm": 0.8105161190032959,
+      "learning_rate": 7.4101244120661105e-06,
+      "loss": 0.4181,
+      "step": 139
+    },
+    {
+      "epoch": 0.3448275862068966,
+      "grad_norm": 2.2261204719543457,
+      "learning_rate": 7.376154443903714e-06,
+      "loss": 0.5089,
+      "step": 140
+    },
+    {
+      "epoch": 0.3472906403940887,
+      "grad_norm": 0.9390556812286377,
+      "learning_rate": 7.342042203498952e-06,
+      "loss": 0.5437,
+      "step": 141
+    },
+    {
+      "epoch": 0.3497536945812808,
+      "grad_norm": 0.8173288702964783,
+      "learning_rate": 7.307789733322146e-06,
+      "loss": 0.4328,
+      "step": 142
+    },
+    {
+      "epoch": 0.3522167487684729,
+      "grad_norm": 1.0669867992401123,
+      "learning_rate": 7.273399084239878e-06,
+      "loss": 0.4557,
+      "step": 143
+    },
+    {
+      "epoch": 0.35467980295566504,
+      "grad_norm": 2.4072816371917725,
+      "learning_rate": 7.238872315392189e-06,
+      "loss": 0.4371,
+      "step": 144
+    },
+    {
+      "epoch": 0.35714285714285715,
+      "grad_norm": 0.8072436451911926,
+      "learning_rate": 7.204211494069292e-06,
+      "loss": 0.3932,
+      "step": 145
+    },
+    {
+      "epoch": 0.35960591133004927,
+      "grad_norm": 1.3799082040786743,
+      "learning_rate": 7.169418695587791e-06,
+      "loss": 0.4526,
+      "step": 146
+    },
+    {
+      "epoch": 0.3620689655172414,
+      "grad_norm": 0.9200634956359863,
+      "learning_rate": 7.134496003166423e-06,
+      "loss": 0.4379,
+      "step": 147
+    },
+    {
+      "epoch": 0.3645320197044335,
+      "grad_norm": 2.47597336769104,
+      "learning_rate": 7.099445507801324e-06,
+      "loss": 0.4638,
+      "step": 148
+    },
+    {
+      "epoch": 0.3669950738916256,
+      "grad_norm": 0.9331015348434448,
+      "learning_rate": 7.06426930814083e-06,
+      "loss": 0.4635,
+      "step": 149
+    },
+    {
+      "epoch": 0.3694581280788177,
+      "grad_norm": 0.8710178732872009,
+      "learning_rate": 7.028969510359821e-06,
+      "loss": 0.3501,
+      "step": 150
+    },
+    {
+      "epoch": 0.37192118226600984,
+      "grad_norm": 1.0825324058532715,
+      "learning_rate": 6.993548228033618e-06,
+      "loss": 0.4367,
+      "step": 151
+    },
+    {
+      "epoch": 0.37438423645320196,
+      "grad_norm": 0.8021703362464905,
+      "learning_rate": 6.9580075820114255e-06,
+      "loss": 0.5561,
+      "step": 152
+    },
+    {
+      "epoch": 0.3768472906403941,
+      "grad_norm": 0.799213707447052,
+      "learning_rate": 6.922349700289348e-06,
+      "loss": 0.4704,
+      "step": 153
+    },
+    {
+      "epoch": 0.3793103448275862,
+      "grad_norm": 1.3380686044692993,
+      "learning_rate": 6.886576717882982e-06,
+      "loss": 0.4525,
+      "step": 154
+    },
+    {
+      "epoch": 0.3817733990147783,
+      "grad_norm": 0.807714581489563,
+      "learning_rate": 6.850690776699574e-06,
+      "loss": 0.3812,
+      "step": 155
+    },
+    {
+      "epoch": 0.3842364532019704,
+      "grad_norm": 0.8078760504722595,
+      "learning_rate": 6.814694025409773e-06,
+      "loss": 0.4816,
+      "step": 156
+    },
+    {
+      "epoch": 0.3866995073891626,
+      "grad_norm": 0.884308934211731,
+      "learning_rate": 6.7785886193189936e-06,
+      "loss": 0.5723,
+      "step": 157
+    },
+    {
+      "epoch": 0.3891625615763547,
+      "grad_norm": 0.8629475831985474,
+      "learning_rate": 6.742376720238346e-06,
+      "loss": 0.4831,
+      "step": 158
+    },
+    {
+      "epoch": 0.3916256157635468,
+      "grad_norm": 0.8618782162666321,
+      "learning_rate": 6.7060604963552125e-06,
+      "loss": 0.4798,
+      "step": 159
+    },
+    {
+      "epoch": 0.39408866995073893,
+      "grad_norm": 0.7644566893577576,
+      "learning_rate": 6.669642122103423e-06,
+      "loss": 0.4965,
+      "step": 160
+    },
+    {
+      "epoch": 0.39655172413793105,
+      "grad_norm": 1.092326283454895,
+      "learning_rate": 6.633123778033061e-06,
+      "loss": 0.4821,
+      "step": 161
+    },
+    {
+      "epoch": 0.39901477832512317,
+      "grad_norm": 0.9640597105026245,
+      "learning_rate": 6.5965076506799e-06,
+      "loss": 0.39,
+      "step": 162
+    },
+    {
+      "epoch": 0.4014778325123153,
+      "grad_norm": 1.1585789918899536,
+      "learning_rate": 6.559795932434489e-06,
+      "loss": 0.4484,
+      "step": 163
+    },
+    {
+      "epoch": 0.4039408866995074,
+      "grad_norm": 0.8530002236366272,
+      "learning_rate": 6.522990821410881e-06,
+      "loss": 0.3783,
+      "step": 164
+    },
+    {
+      "epoch": 0.4064039408866995,
+      "grad_norm": 0.8914724588394165,
+      "learning_rate": 6.486094521315022e-06,
+      "loss": 0.4685,
+      "step": 165
+    },
+    {
+      "epoch": 0.4088669950738916,
+      "grad_norm": 0.7827121019363403,
+      "learning_rate": 6.449109241312803e-06,
+      "loss": 0.4192,
+      "step": 166
+    },
+    {
+      "epoch": 0.41133004926108374,
+      "grad_norm": 1.0978766679763794,
+      "learning_rate": 6.412037195897786e-06,
+      "loss": 0.408,
+      "step": 167
+    },
+    {
+      "epoch": 0.41379310344827586,
+      "grad_norm": 0.8382684588432312,
+      "learning_rate": 6.3748806047586155e-06,
+      "loss": 0.457,
+      "step": 168
+    },
+    {
+      "epoch": 0.41625615763546797,
+      "grad_norm": 1.073023796081543,
+      "learning_rate": 6.337641692646106e-06,
+      "loss": 0.4697,
+      "step": 169
+    },
+    {
+      "epoch": 0.4187192118226601,
+      "grad_norm": 0.9375430941581726,
+      "learning_rate": 6.300322689240042e-06,
+      "loss": 0.4769,
+      "step": 170
+    },
+    {
+      "epoch": 0.4211822660098522,
+      "grad_norm": 0.8222770690917969,
+      "learning_rate": 6.262925829015675e-06,
+      "loss": 0.4652,
+      "step": 171
+    },
+    {
+      "epoch": 0.4236453201970443,
+      "grad_norm": 0.9433236718177795,
+      "learning_rate": 6.2254533511099345e-06,
+      "loss": 0.4507,
+      "step": 172
+    },
+    {
+      "epoch": 0.42610837438423643,
+      "grad_norm": 0.8848790526390076,
+      "learning_rate": 6.187907499187357e-06,
+      "loss": 0.4219,
+      "step": 173
+    },
+    {
+      "epoch": 0.42857142857142855,
+      "grad_norm": 0.8615171313285828,
+      "learning_rate": 6.150290521305746e-06,
+      "loss": 0.4878,
+      "step": 174
+    },
+    {
+      "epoch": 0.43103448275862066,
+      "grad_norm": 0.8783863186836243,
+      "learning_rate": 6.112604669781572e-06,
+      "loss": 0.4351,
+      "step": 175
+    },
+    {
+      "epoch": 0.43349753694581283,
+      "grad_norm": 0.8565014004707336,
+      "learning_rate": 6.074852201055121e-06,
+      "loss": 0.4027,
+      "step": 176
+    },
+    {
+      "epoch": 0.43596059113300495,
+      "grad_norm": 0.8122984766960144,
+      "learning_rate": 6.037035375555376e-06,
+      "loss": 0.4788,
+      "step": 177
+    },
+    {
+      "epoch": 0.43842364532019706,
+      "grad_norm": 0.8158956170082092,
+      "learning_rate": 5.9991564575646855e-06,
+      "loss": 0.4449,
+      "step": 178
+    },
+    {
+      "epoch": 0.4408866995073892,
+      "grad_norm": 0.8699731826782227,
+      "learning_rate": 5.961217715083185e-06,
+      "loss": 0.4864,
+      "step": 179
+    },
+    {
+      "epoch": 0.4433497536945813,
+      "grad_norm": 0.9397453665733337,
+      "learning_rate": 5.923221419693002e-06,
+      "loss": 0.518,
+      "step": 180
+    },
+    {
+      "epoch": 0.4458128078817734,
+      "grad_norm": 0.8958430290222168,
+      "learning_rate": 5.885169846422242e-06,
+      "loss": 0.4862,
+      "step": 181
+    },
+    {
+      "epoch": 0.4482758620689655,
+      "grad_norm": 0.8838856816291809,
+      "learning_rate": 5.847065273608777e-06,
+      "loss": 0.4701,
+      "step": 182
+    },
+    {
+      "epoch": 0.45073891625615764,
+      "grad_norm": 0.8200072050094604,
+      "learning_rate": 5.808909982763825e-06,
+      "loss": 0.3679,
+      "step": 183
+    },
+    {
+      "epoch": 0.45320197044334976,
+      "grad_norm": 0.931821346282959,
+      "learning_rate": 5.770706258435342e-06,
+      "loss": 0.4082,
+      "step": 184
+    },
+    {
+      "epoch": 0.45566502463054187,
+      "grad_norm": 1.0416653156280518,
+      "learning_rate": 5.732456388071247e-06,
+      "loss": 0.4608,
+      "step": 185
+    },
+    {
+      "epoch": 0.458128078817734,
+      "grad_norm": 0.8697349429130554,
+      "learning_rate": 5.6941626618824445e-06,
+      "loss": 0.5462,
+      "step": 186
+    },
+    {
+      "epoch": 0.4605911330049261,
+      "grad_norm": 0.760611355304718,
+      "learning_rate": 5.655827372705712e-06,
+      "loss": 0.4358,
+      "step": 187
+    },
+    {
+      "epoch": 0.4630541871921182,
+      "grad_norm": 0.8112332820892334,
+      "learning_rate": 5.61745281586641e-06,
+      "loss": 0.5072,
+      "step": 188
+    },
+    {
+      "epoch": 0.46551724137931033,
+      "grad_norm": 0.857423722743988,
+      "learning_rate": 5.579041289041045e-06,
+      "loss": 0.4662,
+      "step": 189
+    },
+    {
+      "epoch": 0.46798029556650245,
+      "grad_norm": 0.9442394375801086,
+      "learning_rate": 5.540595092119709e-06,
+      "loss": 0.5124,
+      "step": 190
+    },
+    {
+      "epoch": 0.47044334975369456,
+      "grad_norm": 0.7853634357452393,
+      "learning_rate": 5.502116527068363e-06,
+      "loss": 0.4048,
+      "step": 191
+    },
+    {
+      "epoch": 0.4729064039408867,
+      "grad_norm": 1.0823391675949097,
+      "learning_rate": 5.463607897791006e-06,
+      "loss": 0.4303,
+      "step": 192
+    },
+    {
+      "epoch": 0.4753694581280788,
+      "grad_norm": 1.0174390077590942,
+      "learning_rate": 5.425071509991737e-06,
+      "loss": 0.467,
+      "step": 193
+    },
+    {
+      "epoch": 0.47783251231527096,
+      "grad_norm": 0.9144822955131531,
+      "learning_rate": 5.386509671036695e-06,
+      "loss": 0.4526,
+      "step": 194
+    },
+    {
+      "epoch": 0.4802955665024631,
+      "grad_norm": 0.9582109451293945,
+      "learning_rate": 5.347924689815906e-06,
+      "loss": 0.4019,
+      "step": 195
+    },
+    {
+      "epoch": 0.4827586206896552,
+      "grad_norm": 0.7563319802284241,
+      "learning_rate": 5.309318876605043e-06,
+      "loss": 0.4331,
+      "step": 196
+    },
+    {
+      "epoch": 0.4852216748768473,
+      "grad_norm": 0.8050902485847473,
+      "learning_rate": 5.270694542927089e-06,
+      "loss": 0.4731,
+      "step": 197
+    },
+    {
+      "epoch": 0.4876847290640394,
+      "grad_norm": 0.9717739820480347,
+      "learning_rate": 5.2320540014139405e-06,
+      "loss": 0.4701,
+      "step": 198
+    },
+    {
+      "epoch": 0.49014778325123154,
+      "grad_norm": 0.9432171583175659,
+      "learning_rate": 5.193399565667945e-06,
+      "loss": 0.3581,
+      "step": 199
+    },
+    {
+      "epoch": 0.49261083743842365,
+      "grad_norm": 1.2452778816223145,
+      "learning_rate": 5.154733550123357e-06,
+      "loss": 0.4548,
+      "step": 200
+    },
+    {
+      "epoch": 0.49507389162561577,
+      "grad_norm": 0.78749680519104,
+      "learning_rate": 5.116058269907779e-06,
+      "loss": 0.421,
+      "step": 201
+    },
+    {
+      "epoch": 0.4975369458128079,
+      "grad_norm": 0.8748036026954651,
+      "learning_rate": 5.077376040703533e-06,
+      "loss": 0.4386,
+      "step": 202
+    },
+    {
+      "epoch": 0.5,
+      "grad_norm": 0.8578478693962097,
+      "learning_rate": 5.038689178609011e-06,
+      "loss": 0.5316,
+      "step": 203
+    },
+    {
+      "epoch": 0.5024630541871922,
+      "grad_norm": 0.7721994519233704,
+      "learning_rate": 5e-06,
+      "loss": 0.4235,
+      "step": 204
+    },
+    {
+      "epoch": 0.5049261083743842,
+      "grad_norm": 0.7942141890525818,
+      "learning_rate": 4.96131082139099e-06,
+      "loss": 0.4798,
+      "step": 205
+    },
+    {
+      "epoch": 0.5073891625615764,
+      "grad_norm": 1.1543359756469727,
+      "learning_rate": 4.922623959296469e-06,
+      "loss": 0.4402,
+      "step": 206
+    },
+    {
+      "epoch": 0.5098522167487685,
+      "grad_norm": 0.9413177967071533,
+      "learning_rate": 4.883941730092222e-06,
+      "loss": 0.5872,
+      "step": 207
+    },
+    {
+      "epoch": 0.5123152709359606,
+      "grad_norm": 0.7829416394233704,
+      "learning_rate": 4.845266449876646e-06,
+      "loss": 0.4577,
+      "step": 208
+    },
+    {
+      "epoch": 0.5147783251231527,
+      "grad_norm": 0.7771807312965393,
+      "learning_rate": 4.806600434332056e-06,
+      "loss": 0.4063,
+      "step": 209
+    },
+    {
+      "epoch": 0.5172413793103449,
+      "grad_norm": 0.8723726868629456,
+      "learning_rate": 4.76794599858606e-06,
+      "loss": 0.4745,
+      "step": 210
+    },
+    {
+      "epoch": 0.5197044334975369,
+      "grad_norm": 0.8430385589599609,
+      "learning_rate": 4.729305457072913e-06,
+      "loss": 0.4174,
+      "step": 211
+    },
+    {
+      "epoch": 0.5221674876847291,
+      "grad_norm": 0.6947190165519714,
+      "learning_rate": 4.690681123394959e-06,
+      "loss": 0.378,
+      "step": 212
+    },
+    {
+      "epoch": 0.5246305418719212,
+      "grad_norm": 0.932201623916626,
+      "learning_rate": 4.6520753101840945e-06,
+      "loss": 0.4578,
+      "step": 213
+    },
+    {
+      "epoch": 0.5270935960591133,
+      "grad_norm": 0.8713343739509583,
+      "learning_rate": 4.613490328963307e-06,
+      "loss": 0.39,
+      "step": 214
+    },
+    {
+      "epoch": 0.5295566502463054,
+      "grad_norm": 0.7538230419158936,
+      "learning_rate": 4.574928490008264e-06,
+      "loss": 0.3857,
+      "step": 215
+    },
+    {
+      "epoch": 0.5320197044334976,
+      "grad_norm": 0.8916656970977783,
+      "learning_rate": 4.536392102208998e-06,
+      "loss": 0.5202,
+      "step": 216
+    },
+    {
+      "epoch": 0.5344827586206896,
+      "grad_norm": 0.8922424912452698,
+      "learning_rate": 4.497883472931639e-06,
+      "loss": 0.4081,
+      "step": 217
+    },
+    {
+      "epoch": 0.5369458128078818,
+      "grad_norm": 1.8260999917984009,
+      "learning_rate": 4.459404907880293e-06,
+      "loss": 0.4833,
+      "step": 218
+    },
+    {
+      "epoch": 0.5394088669950738,
+      "grad_norm": 1.0410164594650269,
+      "learning_rate": 4.4209587109589565e-06,
+      "loss": 0.3233,
+      "step": 219
+    },
+    {
+      "epoch": 0.541871921182266,
+      "grad_norm": 0.9134721755981445,
+      "learning_rate": 4.382547184133593e-06,
+      "loss": 0.373,
+      "step": 220
+    },
+    {
+      "epoch": 0.5443349753694581,
+      "grad_norm": 0.8863281607627869,
+      "learning_rate": 4.3441726272942895e-06,
+      "loss": 0.3112,
+      "step": 221
+    },
+    {
+      "epoch": 0.5467980295566502,
+      "grad_norm": 0.7182394862174988,
+      "learning_rate": 4.305837338117557e-06,
+      "loss": 0.4257,
+      "step": 222
+    },
+    {
+      "epoch": 0.5492610837438424,
+      "grad_norm": 1.0430289506912231,
+      "learning_rate": 4.267543611928755e-06,
+      "loss": 0.4123,
+      "step": 223
+    },
+    {
+      "epoch": 0.5517241379310345,
+      "grad_norm": 0.9563698768615723,
+      "learning_rate": 4.229293741564658e-06,
+      "loss": 0.5008,
+      "step": 224
+    },
+    {
+      "epoch": 0.5541871921182266,
+      "grad_norm": 1.4074031114578247,
+      "learning_rate": 4.191090017236177e-06,
+      "loss": 0.4318,
+      "step": 225
+    },
+    {
+      "epoch": 0.5566502463054187,
+      "grad_norm": 0.8157017230987549,
+      "learning_rate": 4.152934726391223e-06,
+      "loss": 0.5214,
+      "step": 226
+    },
+    {
+      "epoch": 0.5591133004926109,
+      "grad_norm": 0.7647298574447632,
+      "learning_rate": 4.114830153577759e-06,
+      "loss": 0.3445,
+      "step": 227
+    },
+    {
+      "epoch": 0.5615763546798029,
+      "grad_norm": 1.0865991115570068,
+      "learning_rate": 4.076778580306999e-06,
+      "loss": 0.3936,
+      "step": 228
+    },
+    {
+      "epoch": 0.5640394088669951,
+      "grad_norm": 1.014061450958252,
+      "learning_rate": 4.0387822849168165e-06,
+      "loss": 0.3716,
+      "step": 229
+    },
+    {
+      "epoch": 0.5665024630541872,
+      "grad_norm": 0.7353817224502563,
+      "learning_rate": 4.000843542435315e-06,
+      "loss": 0.3828,
+      "step": 230
+    },
+    {
+      "epoch": 0.5689655172413793,
+      "grad_norm": 1.005282998085022,
+      "learning_rate": 3.962964624444625e-06,
+      "loss": 0.4572,
+      "step": 231
+    },
+    {
+      "epoch": 0.5714285714285714,
+      "grad_norm": 0.9712240099906921,
+      "learning_rate": 3.92514779894488e-06,
+      "loss": 0.3327,
+      "step": 232
+    },
+    {
+      "epoch": 0.5738916256157636,
+      "grad_norm": 0.7928905487060547,
+      "learning_rate": 3.887395330218429e-06,
+      "loss": 0.3351,
+      "step": 233
+    },
+    {
+      "epoch": 0.5763546798029556,
+      "grad_norm": 0.8094484806060791,
+      "learning_rate": 3.849709478694256e-06,
+      "loss": 0.4305,
+      "step": 234
+    },
+    {
+      "epoch": 0.5788177339901478,
+      "grad_norm": 3.6431384086608887,
+      "learning_rate": 3.8120925008126457e-06,
+      "loss": 0.4964,
+      "step": 235
+    },
+    {
+      "epoch": 0.5812807881773399,
+      "grad_norm": 0.9250912666320801,
+      "learning_rate": 3.7745466488900663e-06,
+      "loss": 0.4114,
+      "step": 236
+    },
+    {
+      "epoch": 0.583743842364532,
+      "grad_norm": 1.0961583852767944,
+      "learning_rate": 3.7370741709843263e-06,
+      "loss": 0.4588,
+      "step": 237
+    },
+    {
+      "epoch": 0.5862068965517241,
+      "grad_norm": 0.857764482498169,
+      "learning_rate": 3.6996773107599605e-06,
+      "loss": 0.3729,
+      "step": 238
+    },
+    {
+      "epoch": 0.5886699507389163,
+      "grad_norm": 1.0068374872207642,
+      "learning_rate": 3.662358307353897e-06,
+      "loss": 0.4843,
+      "step": 239
+    },
+    {
+      "epoch": 0.5911330049261084,
+      "grad_norm": 0.8471587300300598,
+      "learning_rate": 3.6251193952413866e-06,
+      "loss": 0.4503,
+      "step": 240
+    },
+    {
+      "epoch": 0.5935960591133005,
+      "grad_norm": 0.8168319463729858,
+      "learning_rate": 3.587962804102214e-06,
+      "loss": 0.3802,
+      "step": 241
+    },
+    {
+      "epoch": 0.5960591133004927,
+      "grad_norm": 0.8261502981185913,
+      "learning_rate": 3.550890758687199e-06,
+      "loss": 0.4659,
+      "step": 242
+    },
+    {
+      "epoch": 0.5985221674876847,
+      "grad_norm": 1.202130913734436,
+      "learning_rate": 3.5139054786849787e-06,
+      "loss": 0.4397,
+      "step": 243
+    },
+    {
+      "epoch": 0.6009852216748769,
+      "grad_norm": 0.8035978078842163,
+      "learning_rate": 3.4770091785891207e-06,
+      "loss": 0.4111,
+      "step": 244
+    },
+    {
+      "epoch": 0.603448275862069,
+      "grad_norm": 2.1057868003845215,
+      "learning_rate": 3.440204067565511e-06,
+      "loss": 0.4438,
+      "step": 245
+    },
+    {
+      "epoch": 0.6059113300492611,
+      "grad_norm": 0.8185580372810364,
+      "learning_rate": 3.403492349320101e-06,
+      "loss": 0.3776,
+      "step": 246
+    },
+    {
+      "epoch": 0.6083743842364532,
+      "grad_norm": 0.8482059240341187,
+      "learning_rate": 3.3668762219669393e-06,
+      "loss": 0.4184,
+      "step": 247
+    },
+    {
+      "epoch": 0.6108374384236454,
+      "grad_norm": 0.8905600309371948,
+      "learning_rate": 3.330357877896577e-06,
+      "loss": 0.4145,
+      "step": 248
+    },
+    {
+      "epoch": 0.6133004926108374,
+      "grad_norm": 0.7447388172149658,
+      "learning_rate": 3.293939503644788e-06,
+      "loss": 0.4,
+      "step": 249
+    },
+    {
+      "epoch": 0.6157635467980296,
+      "grad_norm": 1.7505443096160889,
+      "learning_rate": 3.2576232797616556e-06,
+      "loss": 0.3793,
+      "step": 250
+    },
+    {
+      "epoch": 0.6182266009852216,
+      "grad_norm": 0.9543902277946472,
+      "learning_rate": 3.2214113806810077e-06,
+      "loss": 0.3999,
+      "step": 251
+    },
+    {
+      "epoch": 0.6206896551724138,
+      "grad_norm": 0.889468789100647,
+      "learning_rate": 3.1853059745902287e-06,
+      "loss": 0.3603,
+      "step": 252
+    },
+    {
+      "epoch": 0.6231527093596059,
+      "grad_norm": 0.9314963817596436,
+      "learning_rate": 3.149309223300428e-06,
+      "loss": 0.4,
+      "step": 253
+    },
+    {
+      "epoch": 0.625615763546798,
+      "grad_norm": 0.7808181047439575,
+      "learning_rate": 3.1134232821170202e-06,
+      "loss": 0.4358,
+      "step": 254
+    },
+    {
+      "epoch": 0.6280788177339901,
+      "grad_norm": 0.8075929284095764,
+      "learning_rate": 3.0776502997106526e-06,
+      "loss": 0.4397,
+      "step": 255
+    },
+    {
+      "epoch": 0.6305418719211823,
+      "grad_norm": 0.991936445236206,
+      "learning_rate": 3.041992417988577e-06,
+      "loss": 0.3703,
+      "step": 256
+    },
+    {
+      "epoch": 0.6330049261083743,
+      "grad_norm": 0.9331387877464294,
+      "learning_rate": 3.0064517719663833e-06,
+      "loss": 0.4161,
+      "step": 257
+    },
+    {
+      "epoch": 0.6354679802955665,
+      "grad_norm": 0.9041942358016968,
+      "learning_rate": 2.9710304896401803e-06,
+      "loss": 0.3642,
+      "step": 258
+    },
+    {
+      "epoch": 0.6379310344827587,
+      "grad_norm": 1.105132818222046,
+      "learning_rate": 2.935730691859172e-06,
+      "loss": 0.3843,
+      "step": 259
+    },
+    {
+      "epoch": 0.6403940886699507,
+      "grad_norm": 0.7953276634216309,
+      "learning_rate": 2.9005544921986774e-06,
+      "loss": 0.3583,
+      "step": 260
+    },
+    {
+      "epoch": 0.6428571428571429,
+      "grad_norm": 0.9104849696159363,
+      "learning_rate": 2.8655039968335774e-06,
+      "loss": 0.4262,
+      "step": 261
+    },
+    {
+      "epoch": 0.645320197044335,
+      "grad_norm": 1.0020502805709839,
+      "learning_rate": 2.83058130441221e-06,
+      "loss": 0.3959,
+      "step": 262
+    },
+    {
+      "epoch": 0.6477832512315271,
+      "grad_norm": 1.0070898532867432,
+      "learning_rate": 2.7957885059307097e-06,
+      "loss": 0.4128,
+      "step": 263
+    },
+    {
+      "epoch": 0.6502463054187192,
+      "grad_norm": 0.8092946410179138,
+      "learning_rate": 2.761127684607811e-06,
+      "loss": 0.4152,
+      "step": 264
+    },
+    {
+      "epoch": 0.6527093596059114,
+      "grad_norm": 0.9570103883743286,
+      "learning_rate": 2.7266009157601226e-06,
+      "loss": 0.4408,
+      "step": 265
+    },
+    {
+      "epoch": 0.6551724137931034,
+      "grad_norm": 0.9353964924812317,
+      "learning_rate": 2.692210266677855e-06,
+      "loss": 0.4548,
+      "step": 266
+    },
+    {
+      "epoch": 0.6576354679802956,
+      "grad_norm": 0.7735761404037476,
+      "learning_rate": 2.65795779650105e-06,
+      "loss": 0.4139,
+      "step": 267
+    },
+    {
+      "epoch": 0.6600985221674877,
+      "grad_norm": 0.8776670098304749,
+      "learning_rate": 2.6238455560962884e-06,
+      "loss": 0.3719,
+      "step": 268
+    },
+    {
+      "epoch": 0.6625615763546798,
+      "grad_norm": 0.9098663330078125,
+      "learning_rate": 2.589875587933892e-06,
+      "loss": 0.4113,
+      "step": 269
+    },
+    {
+      "epoch": 0.6650246305418719,
+      "grad_norm": 3.0552971363067627,
+      "learning_rate": 2.5560499259656323e-06,
+      "loss": 0.4058,
+      "step": 270
+    },
+    {
+      "epoch": 0.6674876847290641,
+      "grad_norm": 0.7983712553977966,
+      "learning_rate": 2.522370595502954e-06,
+      "loss": 0.4385,
+      "step": 271
+    },
+    {
+      "epoch": 0.6699507389162561,
+      "grad_norm": 0.695418119430542,
+      "learning_rate": 2.488839613095695e-06,
+      "loss": 0.3056,
+      "step": 272
+    },
+    {
+      "epoch": 0.6724137931034483,
+      "grad_norm": 1.0092236995697021,
+      "learning_rate": 2.4554589864113566e-06,
+      "loss": 0.3816,
+      "step": 273
+    },
+    {
+      "epoch": 0.6748768472906403,
+      "grad_norm": 0.7949216961860657,
+      "learning_rate": 2.422230714114891e-06,
+      "loss": 0.3537,
+      "step": 274
+    },
+    {
+      "epoch": 0.6773399014778325,
+      "grad_norm": 0.9313486218452454,
+      "learning_rate": 2.3891567857490373e-06,
+      "loss": 0.4997,
+      "step": 275
+    },
+    {
+      "epoch": 0.6798029556650246,
+      "grad_norm": 0.8018554449081421,
+      "learning_rate": 2.3562391816151807e-06,
+      "loss": 0.417,
+      "step": 276
+    },
+    {
+      "epoch": 0.6822660098522167,
+      "grad_norm": 0.9989979267120361,
+      "learning_rate": 2.323479872654805e-06,
+      "loss": 0.4071,
+      "step": 277
+    },
+    {
+      "epoch": 0.6847290640394089,
+      "grad_norm": 0.7565258741378784,
+      "learning_rate": 2.2908808203314637e-06,
+      "loss": 0.3126,
+      "step": 278
+    },
+    {
+      "epoch": 0.687192118226601,
+      "grad_norm": 0.9910544157028198,
+      "learning_rate": 2.2584439765133453e-06,
+      "loss": 0.4361,
+      "step": 279
+    },
+    {
+      "epoch": 0.6896551724137931,
+      "grad_norm": 0.7968178987503052,
+      "learning_rate": 2.226171283356409e-06,
+      "loss": 0.3032,
+      "step": 280
+    },
+    {
+      "epoch": 0.6921182266009852,
+      "grad_norm": 0.8218366503715515,
+      "learning_rate": 2.1940646731880887e-06,
+      "loss": 0.3931,
+      "step": 281
+    },
+    {
+      "epoch": 0.6945812807881774,
+      "grad_norm": 0.8241420984268188,
+      "learning_rate": 2.162126068391601e-06,
+      "loss": 0.4846,
+      "step": 282
+    },
+    {
+      "epoch": 0.6970443349753694,
+      "grad_norm": 0.9964359998703003,
+      "learning_rate": 2.1303573812908383e-06,
+      "loss": 0.4059,
+      "step": 283
+    },
+    {
+      "epoch": 0.6995073891625616,
+      "grad_norm": 0.8862767219543457,
+      "learning_rate": 2.0987605140358823e-06,
+      "loss": 0.387,
+      "step": 284
+    },
+    {
+      "epoch": 0.7019704433497537,
+      "grad_norm": 0.8447444438934326,
+      "learning_rate": 2.0673373584890847e-06,
+      "loss": 0.367,
+      "step": 285
+    },
+    {
+      "epoch": 0.7044334975369458,
+      "grad_norm": 0.719472348690033,
+      "learning_rate": 2.036089796111825e-06,
+      "loss": 0.3834,
+      "step": 286
+    },
+    {
+      "epoch": 0.7068965517241379,
+      "grad_norm": 0.8975826501846313,
+      "learning_rate": 2.0050196978518323e-06,
+      "loss": 0.3538,
+      "step": 287
+    },
+    {
+      "epoch": 0.7093596059113301,
+      "grad_norm": 1.0086376667022705,
+      "learning_rate": 1.9741289240311757e-06,
+      "loss": 0.498,
+      "step": 288
+    },
+    {
+      "epoch": 0.7118226600985221,
+      "grad_norm": 0.9242886304855347,
+      "learning_rate": 1.943419324234871e-06,
+      "loss": 0.442,
+      "step": 289
+    },
+    {
+      "epoch": 0.7142857142857143,
+      "grad_norm": 0.8589749932289124,
+      "learning_rate": 1.9128927372001456e-06,
+      "loss": 0.3315,
+      "step": 290
+    },
+    {
+      "epoch": 0.7167487684729064,
+      "grad_norm": 1.2725121974945068,
+      "learning_rate": 1.8825509907063328e-06,
+      "loss": 0.5441,
+      "step": 291
+    },
+    {
+      "epoch": 0.7192118226600985,
+      "grad_norm": 0.975878119468689,
+      "learning_rate": 1.852395901465441e-06,
+      "loss": 0.4301,
+      "step": 292
+    },
+    {
+      "epoch": 0.7216748768472906,
+      "grad_norm": 0.9230057001113892,
+      "learning_rate": 1.8224292750133743e-06,
+      "loss": 0.4157,
+      "step": 293
+    },
+    {
+      "epoch": 0.7241379310344828,
+      "grad_norm": 0.9983015656471252,
+      "learning_rate": 1.79265290560183e-06,
+      "loss": 0.4511,
+      "step": 294
+    },
+    {
+      "epoch": 0.7266009852216748,
+      "grad_norm": 0.8537778258323669,
+      "learning_rate": 1.7630685760908623e-06,
+      "loss": 0.4724,
+      "step": 295
+    },
+    {
+      "epoch": 0.729064039408867,
+      "grad_norm": 0.8379884958267212,
+      "learning_rate": 1.733678057842142e-06,
+      "loss": 0.3558,
+      "step": 296
+    },
+    {
+      "epoch": 0.7315270935960592,
+      "grad_norm": 0.835649311542511,
+      "learning_rate": 1.7044831106128867e-06,
+      "loss": 0.3725,
+      "step": 297
+    },
+    {
+      "epoch": 0.7339901477832512,
+      "grad_norm": 0.7726622819900513,
+      "learning_rate": 1.675485482450499e-06,
+      "loss": 0.3666,
+      "step": 298
+    },
+    {
+      "epoch": 0.7364532019704434,
+      "grad_norm": 0.9081035256385803,
+      "learning_rate": 1.6466869095879079e-06,
+      "loss": 0.3409,
+      "step": 299
+    },
+    {
+      "epoch": 0.7389162561576355,
+      "grad_norm": 0.7757613658905029,
+      "learning_rate": 1.6180891163396013e-06,
+      "loss": 0.3568,
+      "step": 300
+    },
+    {
+      "epoch": 0.7413793103448276,
+      "grad_norm": 0.9632595777511597,
+      "learning_rate": 1.589693814998391e-06,
+      "loss": 0.4876,
+      "step": 301
+    },
+    {
+      "epoch": 0.7438423645320197,
+      "grad_norm": 0.7990944385528564,
+      "learning_rate": 1.561502705732883e-06,
+      "loss": 0.4158,
+      "step": 302
+    },
+    {
+      "epoch": 0.7463054187192119,
+      "grad_norm": 0.9227884411811829,
+      "learning_rate": 1.533517476485691e-06,
+      "loss": 0.3277,
+      "step": 303
+    },
+    {
+      "epoch": 0.7487684729064039,
+      "grad_norm": 1.0761818885803223,
+      "learning_rate": 1.5057398028723514e-06,
+      "loss": 0.3425,
+      "step": 304
+    },
+    {
+      "epoch": 0.7512315270935961,
+      "grad_norm": 0.7749660015106201,
+      "learning_rate": 1.4781713480810184e-06,
+      "loss": 0.3679,
+      "step": 305
+    },
+    {
+      "epoch": 0.7536945812807881,
+      "grad_norm": 2.2163593769073486,
+      "learning_rate": 1.450813762772863e-06,
+      "loss": 0.407,
+      "step": 306
+    },
+    {
+      "epoch": 0.7561576354679803,
+      "grad_norm": 1.069761037826538,
+      "learning_rate": 1.4236686849832497e-06,
+      "loss": 0.3972,
+      "step": 307
+    },
+    {
+      "epoch": 0.7586206896551724,
+      "grad_norm": 0.9363300800323486,
+      "learning_rate": 1.3967377400236515e-06,
+      "loss": 0.4288,
+      "step": 308
+    },
+    {
+      "epoch": 0.7610837438423645,
+      "grad_norm": 0.8307360410690308,
+      "learning_rate": 1.370022540384347e-06,
+      "loss": 0.5181,
+      "step": 309
+    },
+    {
+      "epoch": 0.7635467980295566,
+      "grad_norm": 0.9606369137763977,
+      "learning_rate": 1.3435246856378524e-06,
+      "loss": 0.5306,
+      "step": 310
+    },
+    {
+      "epoch": 0.7660098522167488,
+      "grad_norm": 0.9734224677085876,
+      "learning_rate": 1.3172457623431706e-06,
+      "loss": 0.53,
+      "step": 311
+    },
+    {
+      "epoch": 0.7684729064039408,
+      "grad_norm": 0.9703786373138428,
+      "learning_rate": 1.2911873439507766e-06,
+      "loss": 0.4361,
+      "step": 312
+    },
+    {
+      "epoch": 0.770935960591133,
+      "grad_norm": 1.0127205848693848,
+      "learning_rate": 1.2653509907084171e-06,
+      "loss": 0.3778,
+      "step": 313
+    },
+    {
+      "epoch": 0.7733990147783252,
+      "grad_norm": 0.6820623874664307,
+      "learning_rate": 1.2397382495676873e-06,
+      "loss": 0.3225,
+      "step": 314
+    },
+    {
+      "epoch": 0.7758620689655172,
+      "grad_norm": 1.9675172567367554,
+      "learning_rate": 1.214350654091413e-06,
+      "loss": 0.4524,
+      "step": 315
+    },
+    {
+      "epoch": 0.7783251231527094,
+      "grad_norm": 0.8468368649482727,
+      "learning_rate": 1.1891897243618184e-06,
+      "loss": 0.4673,
+      "step": 316
+    },
+    {
+      "epoch": 0.7807881773399015,
+      "grad_norm": 1.0746312141418457,
+      "learning_rate": 1.1642569668895171e-06,
+      "loss": 0.4392,
+      "step": 317
+    },
+    {
+      "epoch": 0.7832512315270936,
+      "grad_norm": 1.0559433698654175,
+      "learning_rate": 1.139553874523313e-06,
+      "loss": 0.4199,
+      "step": 318
+    },
+    {
+      "epoch": 0.7857142857142857,
+      "grad_norm": 0.7627314925193787,
+      "learning_rate": 1.1150819263608098e-06,
+      "loss": 0.3582,
+      "step": 319
+    },
+    {
+      "epoch": 0.7881773399014779,
+      "grad_norm": 0.8994935750961304,
+      "learning_rate": 1.0908425876598512e-06,
+      "loss": 0.3603,
+      "step": 320
+    },
+    {
+      "epoch": 0.7906403940886699,
+      "grad_norm": 0.8457913398742676,
+      "learning_rate": 1.0668373097507922e-06,
+      "loss": 0.3291,
+      "step": 321
+    },
+    {
+      "epoch": 0.7931034482758621,
+      "grad_norm": 0.8527902364730835,
+      "learning_rate": 1.0430675299495973e-06,
+      "loss": 0.3809,
+      "step": 322
+    },
+    {
+      "epoch": 0.7955665024630542,
+      "grad_norm": 0.7819089889526367,
+      "learning_rate": 1.0195346714717813e-06,
+      "loss": 0.4148,
+      "step": 323
+    },
+    {
+      "epoch": 0.7980295566502463,
+      "grad_norm": 0.878807544708252,
+      "learning_rate": 9.962401433471985e-07,
+      "loss": 0.3385,
+      "step": 324
+    },
+    {
+      "epoch": 0.8004926108374384,
+      "grad_norm": 1.3636759519577026,
+      "learning_rate": 9.731853403356705e-07,
+      "loss": 0.378,
+      "step": 325
+    },
+    {
+      "epoch": 0.8029556650246306,
+      "grad_norm": 22.09219741821289,
+      "learning_rate": 9.5037164284348e-07,
+      "loss": 0.3528,
+      "step": 326
+    },
+    {
+      "epoch": 0.8054187192118226,
+      "grad_norm": 0.763201117515564,
+      "learning_rate": 9.278004168407151e-07,
+      "loss": 0.4066,
+      "step": 327
+    },
+    {
+      "epoch": 0.8078817733990148,
+      "grad_norm": 0.7910692095756531,
+      "learning_rate": 9.054730137794887e-07,
+      "loss": 0.2703,
+      "step": 328
+    },
+    {
+      "epoch": 0.8103448275862069,
+      "grad_norm": 0.9270244836807251,
+      "learning_rate": 8.833907705130091e-07,
+      "loss": 0.4674,
+      "step": 329
+    },
+    {
+      "epoch": 0.812807881773399,
+      "grad_norm": 0.7657389044761658,
+      "learning_rate": 8.615550092155478e-07,
+      "loss": 0.3766,
+      "step": 330
+    },
+    {
+      "epoch": 0.8152709359605911,
+      "grad_norm": 0.9968824982643127,
+      "learning_rate": 8.399670373032665e-07,
+      "loss": 0.443,
+      "step": 331
+    },
+    {
+      "epoch": 0.8177339901477833,
+      "grad_norm": 0.9094595313072205,
+      "learning_rate": 8.186281473559382e-07,
+      "loss": 0.4102,
+      "step": 332
+    },
+    {
+      "epoch": 0.8201970443349754,
+      "grad_norm": 0.9319184422492981,
+      "learning_rate": 7.975396170395522e-07,
+      "loss": 0.3985,
+      "step": 333
+    },
+    {
+      "epoch": 0.8226600985221675,
+      "grad_norm": 0.8672550916671753,
+      "learning_rate": 7.767027090298207e-07,
+      "loss": 0.4297,
+      "step": 334
+    },
+    {
+      "epoch": 0.8251231527093597,
+      "grad_norm": 1.1561846733093262,
+      "learning_rate": 7.561186709365653e-07,
+      "loss": 0.5102,
+      "step": 335
+    },
+    {
+      "epoch": 0.8275862068965517,
+      "grad_norm": 1.1552369594573975,
+      "learning_rate": 7.357887352290227e-07,
+      "loss": 0.4518,
+      "step": 336
+    },
+    {
+      "epoch": 0.8300492610837439,
+      "grad_norm": 8.381624221801758,
+      "learning_rate": 7.157141191620548e-07,
+      "loss": 0.4467,
+      "step": 337
+    },
+    {
+      "epoch": 0.8325123152709359,
+      "grad_norm": 0.7928099632263184,
+      "learning_rate": 6.958960247032515e-07,
+      "loss": 0.3273,
+      "step": 338
+    },
+    {
+      "epoch": 0.8349753694581281,
+      "grad_norm": 1.1730235815048218,
+      "learning_rate": 6.763356384609809e-07,
+      "loss": 0.4065,
+      "step": 339
+    },
+    {
+      "epoch": 0.8374384236453202,
+      "grad_norm": 1.0426114797592163,
+      "learning_rate": 6.570341316133272e-07,
+      "loss": 0.4301,
+      "step": 340
+    },
+    {
+      "epoch": 0.8399014778325123,
+      "grad_norm": 0.7004266381263733,
+      "learning_rate": 6.379926598379727e-07,
+      "loss": 0.2412,
+      "step": 341
+    },
+    {
+      "epoch": 0.8423645320197044,
+      "grad_norm": 0.8835524320602417,
+      "learning_rate": 6.192123632429986e-07,
+      "loss": 0.4428,
+      "step": 342
+    },
+    {
+      "epoch": 0.8448275862068966,
+      "grad_norm": 0.7402558922767639,
+      "learning_rate": 6.006943662986275e-07,
+      "loss": 0.3081,
+      "step": 343
+    },
+    {
+      "epoch": 0.8472906403940886,
+      "grad_norm": 0.9316273927688599,
+      "learning_rate": 5.824397777698859e-07,
+      "loss": 0.3839,
+      "step": 344
+    },
+    {
+      "epoch": 0.8497536945812808,
+      "grad_norm": 1.0180613994598389,
+      "learning_rate": 5.644496906502233e-07,
+      "loss": 0.3383,
+      "step": 345
+    },
+    {
+      "epoch": 0.8522167487684729,
+      "grad_norm": 0.9619033932685852,
+      "learning_rate": 5.4672518209607e-07,
+      "loss": 0.3643,
+      "step": 346
+    },
+    {
+      "epoch": 0.854679802955665,
+      "grad_norm": 1.3424688577651978,
+      "learning_rate": 5.292673133623372e-07,
+      "loss": 0.5491,
+      "step": 347
+    },
+    {
+      "epoch": 0.8571428571428571,
+      "grad_norm": 1.1439658403396606,
+      "learning_rate": 5.120771297388788e-07,
+      "loss": 0.3623,
+      "step": 348
+    },
+    {
+      "epoch": 0.8596059113300493,
+      "grad_norm": 0.9782744646072388,
+      "learning_rate": 4.951556604879049e-07,
+      "loss": 0.5006,
+      "step": 349
+    },
+    {
+      "epoch": 0.8620689655172413,
+      "grad_norm": 0.9459072351455688,
+      "learning_rate": 4.785039187823503e-07,
+      "loss": 0.4076,
+      "step": 350
+    },
+    {
+      "epoch": 0.8645320197044335,
+      "grad_norm": 0.9256353378295898,
+      "learning_rate": 4.6212290164521554e-07,
+      "loss": 0.4268,
+      "step": 351
+    },
+    {
+      "epoch": 0.8669950738916257,
+      "grad_norm": 0.9616571068763733,
+      "learning_rate": 4.46013589889866e-07,
+      "loss": 0.2967,
+      "step": 352
+    },
+    {
+      "epoch": 0.8694581280788177,
+      "grad_norm": 0.7549436688423157,
+      "learning_rate": 4.3017694806131163e-07,
+      "loss": 0.4012,
+      "step": 353
+    },
+    {
+      "epoch": 0.8719211822660099,
+      "grad_norm": 0.829087495803833,
+      "learning_rate": 4.146139243784475e-07,
+      "loss": 0.3399,
+      "step": 354
+    },
+    {
+      "epoch": 0.874384236453202,
+      "grad_norm": 0.885208785533905,
+      "learning_rate": 3.9932545067728366e-07,
+      "loss": 0.3451,
+      "step": 355
+    },
+    {
+      "epoch": 0.8768472906403941,
+      "grad_norm": 0.9004522562026978,
+      "learning_rate": 3.8431244235515366e-07,
+      "loss": 0.3754,
+      "step": 356
+    },
+    {
+      "epoch": 0.8793103448275862,
+      "grad_norm": 0.9170466065406799,
+      "learning_rate": 3.695757983158954e-07,
+      "loss": 0.3638,
+      "step": 357
+    },
+    {
+      "epoch": 0.8817733990147784,
+      "grad_norm": 1.3507622480392456,
+      "learning_rate": 3.5511640091604293e-07,
+      "loss": 0.3724,
+      "step": 358
+    },
+    {
+      "epoch": 0.8842364532019704,
+      "grad_norm": 0.8250304460525513,
+      "learning_rate": 3.409351159119845e-07,
+      "loss": 0.3516,
+      "step": 359
+    },
+    {
+      "epoch": 0.8866995073891626,
+      "grad_norm": 0.7117695212364197,
+      "learning_rate": 3.270327924081301e-07,
+      "loss": 0.402,
+      "step": 360
+    },
+    {
+      "epoch": 0.8891625615763546,
+      "grad_norm": 1.0209925174713135,
+      "learning_rate": 3.134102628060698e-07,
+      "loss": 0.5029,
+      "step": 361
+    },
+    {
+      "epoch": 0.8916256157635468,
+      "grad_norm": 1.013433814048767,
+      "learning_rate": 3.000683427547374e-07,
+      "loss": 0.4095,
+      "step": 362
+    },
+    {
+      "epoch": 0.8940886699507389,
+      "grad_norm": 0.9493584036827087,
+      "learning_rate": 2.8700783110156507e-07,
+      "loss": 0.4316,
+      "step": 363
+    },
+    {
+      "epoch": 0.896551724137931,
+      "grad_norm": 0.8495911955833435,
+      "learning_rate": 2.742295098446623e-07,
+      "loss": 0.3361,
+      "step": 364
+    },
+    {
+      "epoch": 0.8990147783251231,
+      "grad_norm": 0.8885901570320129,
+      "learning_rate": 2.617341440859883e-07,
+      "loss": 0.4157,
+      "step": 365
+    },
+    {
+      "epoch": 0.9014778325123153,
+      "grad_norm": 1.3665013313293457,
+      "learning_rate": 2.4952248198554075e-07,
+      "loss": 0.4276,
+      "step": 366
+    },
+    {
+      "epoch": 0.9039408866995073,
+      "grad_norm": 0.9520452618598938,
+      "learning_rate": 2.3759525471656163e-07,
+      "loss": 0.3771,
+      "step": 367
+    },
+    {
+      "epoch": 0.9064039408866995,
+      "grad_norm": 0.8242120742797852,
+      "learning_rate": 2.259531764217604e-07,
+      "loss": 0.361,
+      "step": 368
+    },
+    {
+      "epoch": 0.9088669950738916,
+      "grad_norm": 0.9279873371124268,
+      "learning_rate": 2.1459694417055033e-07,
+      "loss": 0.4935,
+      "step": 369
+    },
+    {
+      "epoch": 0.9113300492610837,
+      "grad_norm": 1.1166878938674927,
+      "learning_rate": 2.0352723791731366e-07,
+      "loss": 0.3956,
+      "step": 370
+    },
+    {
+      "epoch": 0.9137931034482759,
+      "grad_norm": 0.8491089344024658,
+      "learning_rate": 1.9274472046068805e-07,
+      "loss": 0.2792,
+      "step": 371
+    },
+    {
+      "epoch": 0.916256157635468,
+      "grad_norm": 1.089476227760315,
+      "learning_rate": 1.8225003740388546e-07,
+      "loss": 0.3604,
+      "step": 372
+    },
+    {
+      "epoch": 0.9187192118226601,
+      "grad_norm": 0.9467745423316956,
+      "learning_rate": 1.7204381711603046e-07,
+      "loss": 0.3956,
+      "step": 373
+    },
+    {
+      "epoch": 0.9211822660098522,
+      "grad_norm": 0.9419111013412476,
+      "learning_rate": 1.621266706945429e-07,
+      "loss": 0.4508,
+      "step": 374
+    },
+    {
+      "epoch": 0.9236453201970444,
+      "grad_norm": 2.099689245223999,
+      "learning_rate": 1.524991919285429e-07,
+      "loss": 0.5749,
+      "step": 375
+    },
+    {
+      "epoch": 0.9261083743842364,
+      "grad_norm": 1.3350390195846558,
+      "learning_rate": 1.431619572633014e-07,
+      "loss": 0.4098,
+      "step": 376
+    },
+    {
+      "epoch": 0.9285714285714286,
+      "grad_norm": 0.912725567817688,
+      "learning_rate": 1.3411552576572562e-07,
+      "loss": 0.4331,
+      "step": 377
+    },
+    {
+      "epoch": 0.9310344827586207,
+      "grad_norm": 0.9236518144607544,
+      "learning_rate": 1.253604390908819e-07,
+      "loss": 0.3197,
+      "step": 378
+    },
+    {
+      "epoch": 0.9334975369458128,
+      "grad_norm": 0.9659357070922852,
+      "learning_rate": 1.1689722144956672e-07,
+      "loss": 0.4172,
+      "step": 379
+    },
+    {
+      "epoch": 0.9359605911330049,
+      "grad_norm": 0.7131395936012268,
+      "learning_rate": 1.0872637957691834e-07,
+      "loss": 0.3812,
+      "step": 380
+    },
+    {
+      "epoch": 0.9384236453201971,
+      "grad_norm": 1.1041388511657715,
+      "learning_rate": 1.008484027020773e-07,
+      "loss": 0.4566,
+      "step": 381
+    },
+    {
+      "epoch": 0.9408866995073891,
+      "grad_norm": 0.7971521615982056,
+      "learning_rate": 9.326376251889202e-08,
+      "loss": 0.343,
+      "step": 382
+    },
+    {
+      "epoch": 0.9433497536945813,
+      "grad_norm": 0.7595178484916687,
+      "learning_rate": 8.597291315767808e-08,
+      "loss": 0.37,
+      "step": 383
+    },
+    {
+      "epoch": 0.9458128078817734,
+      "grad_norm": 0.709429144859314,
+      "learning_rate": 7.897629115802553e-08,
+      "loss": 0.3152,
+      "step": 384
+    },
+    {
+      "epoch": 0.9482758620689655,
+      "grad_norm": 9.246243476867676,
+      "learning_rate": 7.227431544266194e-08,
+      "loss": 0.6465,
+      "step": 385
+    },
+    {
+      "epoch": 0.9507389162561576,
+      "grad_norm": 1.1263999938964844,
+      "learning_rate": 6.58673872923693e-08,
+      "loss": 0.33,
+      "step": 386
+    },
+    {
+      "epoch": 0.9532019704433498,
+      "grad_norm": 0.830968976020813,
+      "learning_rate": 5.97558903219575e-08,
+      "loss": 0.369,
+      "step": 387
+    },
+    {
+      "epoch": 0.9556650246305419,
+      "grad_norm": 0.7675350904464722,
+      "learning_rate": 5.3940190457294486e-08,
+      "loss": 0.3185,
+      "step": 388
+    },
+    {
+      "epoch": 0.958128078817734,
+      "grad_norm": 1.0136815309524536,
+      "learning_rate": 4.842063591339763e-08,
+      "loss": 0.4651,
+      "step": 389
+    },
+    {
+      "epoch": 0.9605911330049262,
+      "grad_norm": 0.9319408535957336,
+      "learning_rate": 4.3197557173584317e-08,
+      "loss": 0.3876,
+      "step": 390
+    },
+    {
+      "epoch": 0.9630541871921182,
+      "grad_norm": 0.8962835669517517,
+      "learning_rate": 3.82712669696822e-08,
+      "loss": 0.4232,
+      "step": 391
+    },
+    {
+      "epoch": 0.9655172413793104,
+      "grad_norm": 0.8527126312255859,
+      "learning_rate": 3.364206026330752e-08,
+      "loss": 0.3564,
+      "step": 392
+    },
+    {
+      "epoch": 0.9679802955665024,
+      "grad_norm": 0.8836209177970886,
+      "learning_rate": 2.9310214228202016e-08,
+      "loss": 0.3616,
+      "step": 393
+    },
+    {
+      "epoch": 0.9704433497536946,
+      "grad_norm": 0.8913058638572693,
+      "learning_rate": 2.527598823363786e-08,
+      "loss": 0.4859,
+      "step": 394
+    },
+    {
+      "epoch": 0.9729064039408867,
+      "grad_norm": 0.8344419598579407,
+      "learning_rate": 2.153962382888841e-08,
+      "loss": 0.3994,
+      "step": 395
+    },
+    {
+      "epoch": 0.9753694581280788,
+      "grad_norm": 0.9367709755897522,
+      "learning_rate": 1.8101344728764236e-08,
+      "loss": 0.321,
+      "step": 396
+    },
+    {
+      "epoch": 0.9778325123152709,
+      "grad_norm": 0.862615704536438,
+      "learning_rate": 1.496135680021993e-08,
+      "loss": 0.3368,
+      "step": 397
+    },
+    {
+      "epoch": 0.9802955665024631,
+      "grad_norm": 0.8697728514671326,
+      "learning_rate": 1.2119848050025084e-08,
+      "loss": 0.3877,
+      "step": 398
+    },
+    {
+      "epoch": 0.9827586206896551,
+      "grad_norm": 0.985674262046814,
+      "learning_rate": 9.576988613511084e-09,
+      "loss": 0.3868,
+      "step": 399
+    },
+    {
+      "epoch": 0.9852216748768473,
+      "grad_norm": 0.7607033252716064,
+      "learning_rate": 7.332930744380906e-09,
+      "loss": 0.4217,
+      "step": 400
+    },
+    {
+      "epoch": 0.9876847290640394,
+      "grad_norm": 0.9515877962112427,
+      "learning_rate": 5.387808805594752e-09,
+      "loss": 0.417,
+      "step": 401
+    },
+    {
+      "epoch": 0.9901477832512315,
+      "grad_norm": 1.0010173320770264,
+      "learning_rate": 3.741739261324817e-09,
+      "loss": 0.4093,
+      "step": 402
+    },
+    {
+      "epoch": 0.9926108374384236,
+      "grad_norm": 0.6813908219337463,
+      "learning_rate": 2.3948206699819787e-09,
+      "loss": 0.3174,
+      "step": 403
+    },
+    {
+      "epoch": 0.9950738916256158,
+      "grad_norm": 0.9744001626968384,
+      "learning_rate": 1.347133678313295e-09,
+      "loss": 0.3682,
+      "step": 404
+    },
+    {
+      "epoch": 0.9975369458128078,
+      "grad_norm": 0.7549907565116882,
+      "learning_rate": 5.987410165758656e-10,
+      "loss": 0.4171,
+      "step": 405
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 0.7962020635604858,
+      "learning_rate": 1.4968749477872746e-10,
+      "loss": 0.359,
+      "step": 406
+    }
+  ],
+  "logging_steps": 1.0,
+  "max_steps": 406,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 0,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": true
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 8.842746096485663e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}