Synchronizing local compiler cache.

.gitattributes

@@ -7077,3 +7077,5 @@ neuronxcc-2.16.372.0+4a9b2326/MODULE_349901478ff7d1ca3ce4+613edded/model.neff fi
 neuronxcc-2.16.372.0+4a9b2326/MODULE_4a56093d8b6239d969e7+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.16.372.0+4a9b2326/MODULE_a6cfe103ba38d7ced324+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
 neuronxcc-2.16.372.0+4a9b2326/MODULE_d02bc244b8a319b871c1+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.16.372.0+4a9b2326/MODULE_34d371c764b2edea8dcd+613edded/model.neff filter=lfs diff=lfs merge=lfs -text
+neuronxcc-2.16.372.0+4a9b2326/MODULE_3d61798cb258025b2edc+613edded/model.neff filter=lfs diff=lfs merge=lfs -text

neuronxcc-2.16.372.0+4a9b2326/0_REGISTRY/0.1.0.dev1/inference/llama/meta-llama/Meta-Llama-3.1-8B/7f1cfbb1aa86ae8f2d64.json

@@ -0,0 +1 @@
+{"architectures": ["LlamaForCausalLM"], "attention_bias": false, "attention_dropout": 0.0, "bos_token_id": 128000, "eos_token_id": 128001, "head_dim": 128, "hidden_act": "silu", "hidden_size": 4096, "initializer_range": 0.02, "intermediate_size": 14336, "max_position_embeddings": 131072, "mlp_bias": false, "model_type": "llama", "neuron": {"auto_cast_type": "bf16", "batch_size": 1, "checkpoint_id": "meta-llama/Meta-Llama-3.1-8B", "checkpoint_revision": "d04e592bb4f6aa9cfee91e2e20afa771667e1d4b", "compiler_type": "neuronx-cc", "compiler_version": "2.16.372.0+4a9b2326", "num_cores": 2, "sequence_length": 4096, "task": "text-generation"}, "num_attention_heads": 32, "num_hidden_layers": 32, "num_key_value_heads": 8, "pretraining_tp": 1, "rms_norm_eps": 1e-05, "rope_scaling": {"factor": 8.0, "high_freq_factor": 4.0, "low_freq_factor": 1.0, "original_max_position_embeddings": 8192, "rope_type": "llama3"}, "rope_theta": 500000.0, "tie_word_embeddings": false, "torch_dtype": "bfloat16", "use_cache": true, "vocab_size": 128256}

neuronxcc-2.16.372.0+4a9b2326/MODULE_34d371c764b2edea8dcd+613edded/compile_flags.json

@@ -0,0 +1 @@
+["--target=trn1", "--model-type=transformer", "--auto-cast=none", "--execute-repetition=1"]

neuronxcc-2.16.372.0+4a9b2326/MODULE_34d371c764b2edea8dcd+613edded/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b3e44cf86dd1d0e68e104812550ae07601f6173588a9533c6638fc6b8119354
+size 331813

neuronxcc-2.16.372.0+4a9b2326/MODULE_34d371c764b2edea8dcd+613edded/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b2ebf416056af21adc3bc3d71057c68e6353e2f8d7e384b6741366543b8ec0de
+size 7803904

neuronxcc-2.16.372.0+4a9b2326/MODULE_3d61798cb258025b2edc+613edded/compile_flags.json

@@ -0,0 +1 @@
+["--target=trn1", "--model-type=transformer", "--auto-cast=none", "--execute-repetition=1"]

neuronxcc-2.16.372.0+4a9b2326/MODULE_3d61798cb258025b2edc+613edded/model.hlo_module.pb

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:042cfd3f7625001a8965d97f178952ab80e285cd0fe0f299657e2763070f701d
+size 401014

neuronxcc-2.16.372.0+4a9b2326/MODULE_3d61798cb258025b2edc+613edded/model.neff

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e7c98cc9bf2a57760705a932a6f1c62de9f0d31c125e997d8b03d2ae2ad602a
+size 5970944