Update model

adapter_config.json

@@ -1,7 +1,7 @@
 {
   "alpha_pattern": {},
   "auto_mapping": null,
-  "base_model_name_or_path": "Isaachhe/phi-2_dev",
+  "base_model_name_or_path": "microsoft/phi-2",
   "bias": "none",
   "fan_in_fan_out": false,
   "inference_mode": true,
@@ -19,8 +19,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "query_key_value",
     "fc1",
+    "q_proj",
+    "k_proj",
+    "v_proj",
     "dense",
     "fc2"
   ],

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29d0e6e9a71dd29aece673043fd727fb772dd00815cb82f0a4f0f7e503797942
-size 335579120
+oid sha256:38ba97037e32366eb3a5b831aa0807a533b4ba56975b49d314eaf793665f768a
+size 377539280

config.json

@@ -4,6 +4,10 @@
     "PhiForCausalLM"
   ],
   "attention_dropout": 0.0,
+  "auto_map": {
+    "AutoConfig": "configuration_phi.PhiConfig",
+    "AutoModelForCausalLM": "modeling_phi.PhiForCausalLM"
+  },
   "bos_token_id": 50256,
   "embd_pdrop": 0.0,
   "eos_token_id": 50256,
@@ -22,8 +26,8 @@
   "model_type": "bunny-phi",
   "num_attention_heads": 32,
   "num_hidden_layers": 32,
+  "num_key_value_heads": 32,
   "partial_rotary_factor": 0.4,
-  "pretraining_tp": 1,
   "qk_layernorm": false,
   "resid_pdrop": 0.1,
   "rope_scaling": null,

non_lora_trainables.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fd0abfe41d1ffce799130123cfda6aad65743cf6cb9973f4dddfa111a3196588
+oid sha256:17ba166919e7778444751a48b78efae3278f7320da8d50b7e325adf0ac153548
 size 18362544