Hugging Face's logo

taesiri
/
LLaVA-ImageDiff

Model card Files Community
taesiri commited on
Commit
08f0b03
•
1 Parent(s): ec4c6d4

initial commit

Browse files
Files changed (6) hide show
  1. checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/README.md +9 -0
  2. checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/adapter_config.json +26 -0
  3. checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/adapter_model.bin +3 -0
  4. checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/config.json +49 -0
  5. checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/non_lora_trainables.bin +3 -0
  6. checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/trainer_state.json +0 -0
checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/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
checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "auto_mapping": null,
3
+ "base_model_name_or_path": "liuhaotian/llava-v1.5-13b",
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
+ "up_proj",
18
+ "v_proj",
19
+ "down_proj",
20
+ "k_proj",
21
+ "q_proj",
22
+ "gate_proj",
23
+ "o_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/adapter_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:bf0e3aa74f2749556f970a6c65bb63f4b3c89ad1f8a4135c27e47defac18de01
3
+ size 1001584778
checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/config.json ADDED
@@ -0,0 +1,49 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "liuhaotian/llava-v1.5-13b",
3
+ "architectures": [
4
+ "LlavaLlamaForCausalLM"
5
+ ],
6
+ "attention_bias": false,
7
+ "attention_dropout": 0.0,
8
+ "bos_token_id": 1,
9
+ "eos_token_id": 2,
10
+ "freeze_mm_mlp_adapter": false,
11
+ "freeze_mm_vision_resampler": false,
12
+ "hidden_act": "silu",
13
+ "hidden_size": 5120,
14
+ "image_aspect_ratio": "pad",
15
+ "initializer_range": 0.02,
16
+ "intermediate_size": 13824,
17
+ "max_length": 4096,
18
+ "max_position_embeddings": 4096,
19
+ "mm_hidden_size": 1024,
20
+ "mm_patch_merge_type": "flat",
21
+ "mm_projector_lr": 2e-05,
22
+ "mm_projector_type": "mlp2x_gelu",
23
+ "mm_resampler_type": null,
24
+ "mm_use_im_patch_token": false,
25
+ "mm_use_im_start_end": false,
26
+ "mm_vision_select_feature": "patch",
27
+ "mm_vision_select_layer": -2,
28
+ "mm_vision_tower": "openai/clip-vit-large-patch14-336",
29
+ "model_type": "llava_llama",
30
+ "num_attention_heads": 40,
31
+ "num_hidden_layers": 40,
32
+ "num_key_value_heads": 40,
33
+ "pad_token_id": 0,
34
+ "pretraining_tp": 1,
35
+ "rms_norm_eps": 1e-05,
36
+ "rope_scaling": null,
37
+ "rope_theta": 10000.0,
38
+ "tie_word_embeddings": false,
39
+ "tokenizer_model_max_length": 2048,
40
+ "tokenizer_padding_side": "right",
41
+ "torch_dtype": "float16",
42
+ "transformers_version": "4.37.2",
43
+ "tune_mm_mlp_adapter": false,
44
+ "tune_mm_vision_resampler": false,
45
+ "unfreeze_mm_vision_tower": false,
46
+ "use_cache": true,
47
+ "use_mm_proj": true,
48
+ "vocab_size": 32000
49
+ }
checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/non_lora_trainables.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:edc9923e9d4e7f2394fe61acb69e2b1e637be2f3195cabb8a6e7b6f6a5260a3a
3
+ size 62937264
checkpoints/llava-v1.5-13b-task-lora-diff-caption-2/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff