Upload PhiForCausalLM

config.json

@@ -4,17 +4,25 @@
   "architectures": [
     "PhiForCausalLM"
   ],
+  "attention_dropout": 0.0,
   "attn_pdrop": 0.0,
   "auto_map": {
     "AutoConfig": "microsoft/phi-1_5--configuration_phi.PhiConfig",
     "AutoModelForCausalLM": "microsoft/phi-1_5--modeling_phi.PhiForCausalLM"
   },
+  "bos_token_id": 1,
   "embd_pdrop": 0.0,
+  "eos_token_id": 2,
   "flash_attn": false,
   "flash_rotary": false,
   "fused_dense": false,
+  "hidden_act": "gelu_new",
+  "hidden_size": 2048,
   "initializer_range": 0.02,
+  "intermediate_size": 8192,
+  "layer_norm_eps": 1e-05,
   "layer_norm_epsilon": 1e-05,
+  "max_position_embeddings": 2048,
   "model_type": "phi",
   "n_embd": 2048,
   "n_head": 32,
@@ -22,10 +30,17 @@
   "n_inner": null,
   "n_layer": 24,
   "n_positions": 2048,
+  "num_attention_heads": 32,
+  "num_hidden_layers": 24,
+  "partial_rotary_factor": 0.5,
+  "qk_layernorm": false,
   "resid_pdrop": 0.0,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
   "rotary_dim": 32,
   "tie_word_embeddings": false,
-  "torch_dtype": "bfloat16",
-  "transformers_version": "4.35.2",
+  "torch_dtype": "float16",
+  "transformers_version": "4.36.0",
+  "use_cache": true,
   "vocab_size": 51200
 }

generation_config.json

@@ -1,4 +1,4 @@
 {
   "_from_model_config": true,
-  "transformers_version": "4.35.2"
+  "transformers_version": "4.36.0"
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8f175cd5282f0055a6925ea500990e1a0b4cd58e823cb9f13125222650537644
-size 2836567400
+oid sha256:2354112303f7cda7458f24f27ed4d7728ada93adf0b8ba17a73d3ce6fafab35a
+size 2836568312