Update config.json

config.json

@@ -4,11 +4,10 @@
   "architectures": [
     "PhiForCausalLM"
   ],
-  "attention_dropout": 0.0,
   "attn_pdrop": 0.0,
   "auto_map": {
-    "AutoConfig": "rhysjones/phi-2-orange--configuration_phi.PhiConfig",
-    "AutoModelForCausalLM": "rhysjones/phi-2-orange--modeling_phi.PhiForCausalLM"
+    "AutoConfig": "configuration_phi.PhiConfig",
+    "AutoModelForCausalLM": "modeling_phi.PhiForCausalLM"
   },
   "bos_token_id": null,
   "embd_pdrop": 0.0,
@@ -16,14 +15,15 @@
   "flash_attn": false,
   "flash_rotary": false,
   "fused_dense": false,
-  "hidden_act": "silu",
-  "hidden_size": 4096,
+  "hidden_act": "gelu_new",
+  "hidden_size": 2560,
   "img_processor": null,
   "initializer_range": 0.02,
-  "intermediate_size": 14336,
+  "intermediate_size": 8192,
+  "layer_norm_eps": 1e-05,
   "layer_norm_epsilon": 1e-05,
   "max_position_embeddings": 2048,
-  "model_type": "mixtral",
+  "model_type": "phi-msft",
   "n_embd": 2560,
   "n_head": 32,
   "n_head_kv": null,
@@ -32,19 +32,21 @@
   "n_positions": 2048,
   "num_attention_heads": 32,
   "num_experts_per_tok": 2,
-  "num_hidden_layers": 32,
-  "num_key_value_heads": 8,
+  "num_hidden_layers": 24,
+  "num_experts_per_tok": 2,
+  "num_local_experts": 4,
+  "num_key_value_heads": 32,
   "num_local_experts": 2,
-  "output_router_logits": false,
+  "partial_rotary_factor": 0.5,
+  "qk_layernorm": false,
   "resid_pdrop": 0.1,
-  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
   "rope_theta": 10000.0,
   "rotary_dim": 32,
-  "router_aux_loss_coef": 0.001,
   "sliding_window": null,
   "tie_word_embeddings": false,
   "torch_dtype": "float16",
-  "transformers_version": "4.37.2",
+  "transformers_version": "4.37.1",
   "use_cache": false,
   "vocab_size": 51200
-}
+}