cdh commited on
Commit
ecdc73f
1 Parent(s): 9524023

Upload 6 files

Browse files
README.md CHANGED
@@ -1,3 +1,9 @@
1
  ---
2
- license: apache-2.0
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": "/home/d.croce/Llama-2-13b-chat-hf",
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": 16,
11
+ "lora_dropout": 0.05,
12
+ "modules_to_save": null,
13
+ "peft_type": "LORA",
14
+ "r": 64,
15
+ "revision": null,
16
+ "target_modules": [
17
+ "gate_proj",
18
+ "k_proj",
19
+ "down_proj",
20
+ "v_proj",
21
+ "up_proj",
22
+ "o_proj",
23
+ "q_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:f61c16e2535d53c62e97035a80dfe9ff1db9a5d4f1d796687d7616b37aa32f77
3
+ size 500889293
config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "/home/d.croce/Llama-2-13b-chat-hf",
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": "square",
12
+ "image_grid_pinpoints": null,
13
+ "initializer_range": 0.02,
14
+ "intermediate_size": 13824,
15
+ "max_position_embeddings": 4096,
16
+ "mm_hidden_size": 1024,
17
+ "mm_use_im_patch_token": false,
18
+ "mm_use_im_start_end": false,
19
+ "mm_vision_select_feature": "patch",
20
+ "mm_vision_select_layer": -2,
21
+ "mm_vision_tower": "openai/clip-vit-large-patch14",
22
+ "model_type": "llava",
23
+ "num_attention_heads": 40,
24
+ "num_hidden_layers": 40,
25
+ "num_key_value_heads": 40,
26
+ "pad_token_id": 0,
27
+ "pretraining_tp": 1,
28
+ "rms_norm_eps": 1e-05,
29
+ "rope_scaling": null,
30
+ "tie_word_embeddings": false,
31
+ "torch_dtype": "float16",
32
+ "transformers_version": "4.31.0",
33
+ "tune_mm_mlp_adapter": false,
34
+ "use_cache": true,
35
+ "use_mm_proj": true,
36
+ "vocab_size": 32000
37
+ }
non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bfd21833a7aca206151c5761e815adf5ee5b443fc79231d8beb206ef39c8118f
3
+ size 10497175
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff