yuzhang commited on
Commit
cd01d3a
1 Parent(s): 97a4a43

Upload 6 files

Browse files
README.md CHANGED
@@ -1,3 +1,9 @@
1
  ---
2
- license: mit
3
  ---
 
 
 
 
 
 
 
1
  ---
2
+ library_name: peft
3
  ---
4
+ ## Training procedure
5
+
6
+ ### Framework versions
7
+
8
+
9
+ - PEFT 0.4.0
adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "lmsys/vicuna-13b-v1.5",
4
+ "bias": "none",
5
+ "fan_in_fan_out": false,
6
+ "inference_mode": true,
7
+ "init_lora_weights": true,
8
+ "layers_pattern": null,
9
+ "layers_to_transform": null,
10
+ "lora_alpha": 256,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 128,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "o_proj",
18
+ "v_proj",
19
+ "q_proj",
20
+ "k_proj",
21
+ "up_proj",
22
+ "gate_proj",
23
+ "down_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1570554eca78912f0f450065b549bd3b388e59480479f167708c87b45e72d531
3
+ size 1001584333
config.json ADDED
@@ -0,0 +1,41 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "lmsys/vicuna-13b-v1.5",
3
+ "architectures": [
4
+ "LlamaForCausalLM"
5
+ ],
6
+ "bos_token_id": 1,
7
+ "eos_token_id": 2,
8
+ "freeze_mm_mlp_adapter": false,
9
+ "hidden_act": "silu",
10
+ "hidden_size": 5120,
11
+ "image_aspect_ratio": "pad",
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 13824,
14
+ "max_length": 4096,
15
+ "max_position_embeddings": 4096,
16
+ "mm_hidden_size": 1024,
17
+ "mm_projector_lr": 2e-05,
18
+ "mm_projector_type": "mlp2x_gelu",
19
+ "mm_use_im_patch_token": false,
20
+ "mm_use_im_start_end": false,
21
+ "mm_vision_select_feature": "patch",
22
+ "mm_vision_select_layer": -2,
23
+ "mm_vision_tower": "openai/clip-vit-large-patch14-336",
24
+ "model_type": "llava",
25
+ "num_attention_heads": 40,
26
+ "num_hidden_layers": 40,
27
+ "num_key_value_heads": 40,
28
+ "pad_token_id": 0,
29
+ "pretraining_tp": 1,
30
+ "rms_norm_eps": 1e-05,
31
+ "rope_scaling": null,
32
+ "tie_word_embeddings": false,
33
+ "tokenizer_model_max_length": 2048,
34
+ "tokenizer_padding_side": "right",
35
+ "torch_dtype": "float16",
36
+ "transformers_version": "4.31.0",
37
+ "tune_mm_mlp_adapter": false,
38
+ "use_cache": true,
39
+ "use_mm_proj": true,
40
+ "vocab_size": 32000
41
+ }
non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e9da06335567ecbe39533323239430ea8518d2c8900faca3259541b270926473
3
+ size 62936807
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff