BoyaWu10 commited on
Commit
ef8c09c
1 Parent(s): 0f84d3e

Update model

Browse files
adapter_config.json CHANGED
@@ -1,7 +1,7 @@
1
  {
2
  "alpha_pattern": {},
3
  "auto_mapping": null,
4
- "base_model_name_or_path": "Isaachhe/phi-2_dev",
5
  "bias": "none",
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
@@ -22,7 +22,9 @@
22
  "fc2",
23
  "dense",
24
  "fc1",
25
- "query_key_value"
 
 
26
  ],
27
  "task_type": "CAUSAL_LM"
28
  }
 
1
  {
2
  "alpha_pattern": {},
3
  "auto_mapping": null,
4
+ "base_model_name_or_path": "microsoft/phi-2",
5
  "bias": "none",
6
  "fan_in_fan_out": false,
7
  "inference_mode": true,
 
22
  "fc2",
23
  "dense",
24
  "fc1",
25
+ "v_proj",
26
+ "k_proj",
27
+ "q_proj"
28
  ],
29
  "task_type": "CAUSAL_LM"
30
  }
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7ed26367984c3d2a7996ca30cdc45645ca80d37007001252dd6049b7f3bd1f08
3
- size 335579120
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2d1284419bd2a331515eb0a17e99d781cabe2198407f8a6c98e66fdc105ab303
3
+ size 377539280
config.json CHANGED
@@ -4,6 +4,10 @@
4
  "PhiForCausalLM"
5
  ],
6
  "attention_dropout": 0.0,
 
 
 
 
7
  "bos_token_id": 50256,
8
  "embd_pdrop": 0.0,
9
  "eos_token_id": 50256,
@@ -18,13 +22,12 @@
18
  "mm_hidden_size": 1152,
19
  "mm_projector_lr": 2e-05,
20
  "mm_projector_type": "mlp2x_gelu",
21
- "mm_vision_select_feature": "patch",
22
  "mm_vision_tower": "google/siglip-so400m-patch14-384",
23
  "model_type": "bunny-phi",
24
  "num_attention_heads": 32,
25
  "num_hidden_layers": 32,
 
26
  "partial_rotary_factor": 0.4,
27
- "pretraining_tp": 1,
28
  "qk_layernorm": false,
29
  "resid_pdrop": 0.1,
30
  "rope_scaling": null,
 
4
  "PhiForCausalLM"
5
  ],
6
  "attention_dropout": 0.0,
7
+ "auto_map": {
8
+ "AutoConfig": "configuration_phi.PhiConfig",
9
+ "AutoModelForCausalLM": "modeling_phi.PhiForCausalLM"
10
+ },
11
  "bos_token_id": 50256,
12
  "embd_pdrop": 0.0,
13
  "eos_token_id": 50256,
 
22
  "mm_hidden_size": 1152,
23
  "mm_projector_lr": 2e-05,
24
  "mm_projector_type": "mlp2x_gelu",
 
25
  "mm_vision_tower": "google/siglip-so400m-patch14-384",
26
  "model_type": "bunny-phi",
27
  "num_attention_heads": 32,
28
  "num_hidden_layers": 32,
29
+ "num_key_value_heads": 32,
30
  "partial_rotary_factor": 0.4,
 
31
  "qk_layernorm": false,
32
  "resid_pdrop": 0.1,
33
  "rope_scaling": null,
non_lora_trainables.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e9159cdbf2ec303e5929e7b46561607ebaecc337fcde60670351a8d047fb2b5b
3
  size 19017904
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:04ab1782b83594fb2404402b5b42ab2bbb382e451389c930a5961b26b36a29af
3
  size 19017904
trainer_state.json CHANGED
The diff for this file is too large to render. See raw diff