yaolily commited on
Commit
b48a33c
1 Parent(s): 0011b1a

Upload 4 files

Browse files

upload config files

Files changed (4) hide show
  1. README.md +9 -0
  2. adapter_config.json +26 -0
  3. config.json +40 -0
  4. trainer_state.json +0 -0
README.md ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
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": "/home/shared/LargeLanguageModels/vicuna/vicuna-7b-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
+ "k_proj",
19
+ "q_proj",
20
+ "down_proj",
21
+ "up_proj",
22
+ "gate_proj",
23
+ "v_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/shared/LargeLanguageModels/vicuna/vicuna-7b-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": 4096,
11
+ "image_aspect_ratio": "pad",
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 11008,
14
+ "max_position_embeddings": 4096,
15
+ "mm_hidden_size": 1024,
16
+ "mm_projector_lr": 2e-05,
17
+ "mm_projector_type": "dabstractor2_144",
18
+ "mm_use_im_patch_token": false,
19
+ "mm_use_im_start_end": false,
20
+ "mm_vision_select_feature": "patch",
21
+ "mm_vision_select_layer": -2,
22
+ "mm_vision_tower": "/home/yaolinli/code/official_ckpts/clip-vit-large-patch14-336",
23
+ "model_type": "llava",
24
+ "num_attention_heads": 32,
25
+ "num_hidden_layers": 32,
26
+ "num_key_value_heads": 32,
27
+ "pad_token_id": 0,
28
+ "pretraining_tp": 1,
29
+ "rms_norm_eps": 1e-05,
30
+ "rope_scaling": null,
31
+ "tie_word_embeddings": false,
32
+ "tokenizer_model_max_length": 2048,
33
+ "tokenizer_padding_side": "right",
34
+ "torch_dtype": "float16",
35
+ "transformers_version": "4.31.0",
36
+ "tune_mm_mlp_adapter": false,
37
+ "use_cache": true,
38
+ "use_mm_proj": true,
39
+ "vocab_size": 32000
40
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff