Hugging Face's logo

A2Amir
/
SF_A68_LLAVA_IDL_SFT

Transformers
Safetensors
trl
sft
Inference Endpoints
Model card Files Community
A2Amir commited on
Commit
2152425
·
verified ·
1 Parent(s): 0703a5d

Training in progress, step 500

Browse files
Files changed (6) hide show
  1. adapter_config.json +180 -0
  2. adapter_model.safetensors +3 -0
  3. special_tokens_map.json +30 -0
  4. tokenizer.json +0 -0
  5. tokenizer_config.json +68 -0
  6. training_args.bin +3 -0
adapter_config.json ADDED
@@ -0,0 +1,180 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": {
4
+ "base_model_class": "LlavaNextForConditionalGeneration",
5
+ "parent_library": "transformers.models.llava_next.modeling_llava_next"
6
+ },
7
+ "base_model_name_or_path": "llava-hf/llava-v1.6-mistral-7b-hf",
8
+ "bias": "none",
9
+ "fan_in_fan_out": false,
10
+ "inference_mode": true,
11
+ "init_lora_weights": "gaussian",
12
+ "layer_replication": null,
13
+ "layers_pattern": null,
14
+ "layers_to_transform": null,
15
+ "loftq_config": {},
16
+ "lora_alpha": 64,
17
+ "lora_dropout": 0.1,
18
+ "megatron_config": null,
19
+ "megatron_core": "megatron.core",
20
+ "modules_to_save": null,
21
+ "peft_type": "LORA",
22
+ "r": 128,
23
+ "rank_pattern": {},
24
+ "revision": null,
25
+ "target_modules": [
26
+ "vision_tower.vision_model.encoder.layers.6.self_attn.out_proj",
27
+ "vision_tower.vision_model.encoder.layers.12.self_attn.k_proj",
28
+ "vision_tower.vision_model.encoder.layers.3.self_attn.k_proj",
29
+ "vision_tower.vision_model.encoder.layers.2.self_attn.v_proj",
30
+ "vision_tower.vision_model.encoder.layers.5.self_attn.q_proj",
31
+ "vision_tower.vision_model.encoder.layers.16.mlp.fc1",
32
+ "vision_tower.vision_model.encoder.layers.18.mlp.fc2",
33
+ "vision_tower.vision_model.encoder.layers.3.mlp.fc1",
34
+ "vision_tower.vision_model.encoder.layers.8.mlp.fc2",
35
+ "vision_tower.vision_model.encoder.layers.17.mlp.fc2",
36
+ "vision_tower.vision_model.encoder.layers.9.self_attn.q_proj",
37
+ "vision_tower.vision_model.encoder.layers.0.self_attn.k_proj",
38
+ "vision_tower.vision_model.encoder.layers.5.self_attn.k_proj",
39
+ "vision_tower.vision_model.encoder.layers.15.mlp.fc2",
40
+ "vision_tower.vision_model.encoder.layers.7.self_attn.q_proj",
41
+ "vision_tower.vision_model.encoder.layers.22.self_attn.out_proj",
42
+ "vision_tower.vision_model.encoder.layers.20.self_attn.k_proj",
43
+ "vision_tower.vision_model.encoder.layers.8.mlp.fc1",
44
+ "vision_tower.vision_model.encoder.layers.3.mlp.fc2",
45
+ "vision_tower.vision_model.encoder.layers.13.self_attn.q_proj",
46
+ "vision_tower.vision_model.encoder.layers.19.mlp.fc2",
47
+ "vision_tower.vision_model.encoder.layers.19.self_attn.k_proj",
48
+ "vision_tower.vision_model.encoder.layers.15.self_attn.out_proj",
49
+ "vision_tower.vision_model.encoder.layers.22.self_attn.v_proj",
50
+ "vision_tower.vision_model.encoder.layers.21.mlp.fc1",
51
+ "vision_tower.vision_model.encoder.layers.14.self_attn.q_proj",
52
+ "vision_tower.vision_model.encoder.layers.8.self_attn.v_proj",
53
+ "vision_tower.vision_model.encoder.layers.17.self_attn.out_proj",
54
+ "vision_tower.vision_model.encoder.layers.23.self_attn.v_proj",
55
+ "vision_tower.vision_model.encoder.layers.16.self_attn.v_proj",
56
+ "vision_tower.vision_model.encoder.layers.9.mlp.fc2",
57
+ "vision_tower.vision_model.encoder.layers.5.mlp.fc1",
58
+ "vision_tower.vision_model.encoder.layers.0.mlp.fc2",
59
+ "vision_tower.vision_model.encoder.layers.11.mlp.fc1",
60
+ "vision_tower.vision_model.encoder.layers.0.self_attn.v_proj",
61
+ "vision_tower.vision_model.encoder.layers.4.self_attn.v_proj",
62
+ "vision_tower.vision_model.encoder.layers.21.self_attn.k_proj",
63
+ "vision_tower.vision_model.encoder.layers.20.mlp.fc2",
64
+ "vision_tower.vision_model.encoder.layers.13.self_attn.v_proj",
65
+ "vision_tower.vision_model.encoder.layers.15.self_attn.q_proj",
66
+ "vision_tower.vision_model.encoder.layers.14.self_attn.v_proj",
67
+ "vision_tower.vision_model.encoder.layers.19.self_attn.q_proj",
68
+ "vision_tower.vision_model.encoder.layers.12.self_attn.v_proj",
69
+ "vision_tower.vision_model.encoder.layers.18.self_attn.q_proj",
70
+ "vision_tower.vision_model.encoder.layers.4.mlp.fc1",
71
+ "vision_tower.vision_model.encoder.layers.6.mlp.fc1",
72
+ "multi_modal_projector.linear_1.lora_A.default",
73
+ "vision_tower.vision_model.encoder.layers.7.mlp.fc1",
74
+ "vision_tower.vision_model.encoder.layers.7.self_attn.v_proj",
75
+ "vision_tower.vision_model.encoder.layers.23.mlp.fc1",
76
+ "vision_tower.vision_model.encoder.layers.4.mlp.fc2",
77
+ "vision_tower.vision_model.encoder.layers.21.self_attn.q_proj",
78
+ "vision_tower.vision_model.encoder.layers.2.self_attn.out_proj",
79
+ "vision_tower.vision_model.encoder.layers.19.self_attn.v_proj",
80
+ "vision_tower.vision_model.encoder.layers.8.self_attn.q_proj",
81
+ "vision_tower.vision_model.encoder.layers.12.self_attn.out_proj",
82
+ "vision_tower.vision_model.encoder.layers.10.self_attn.v_proj",
83
+ "vision_tower.vision_model.encoder.layers.6.mlp.fc2",
84
+ "vision_tower.vision_model.encoder.layers.14.self_attn.out_proj",
85
+ "vision_tower.vision_model.encoder.layers.20.self_attn.v_proj",
86
+ "vision_tower.vision_model.encoder.layers.1.mlp.fc2",
87
+ "vision_tower.vision_model.encoder.layers.13.mlp.fc1",
88
+ "vision_tower.vision_model.encoder.layers.17.self_attn.k_proj",
89
+ "vision_tower.vision_model.encoder.layers.3.self_attn.q_proj",
90
+ "vision_tower.vision_model.encoder.layers.13.mlp.fc2",
91
+ "vision_tower.vision_model.encoder.layers.9.self_attn.v_proj",
92
+ "vision_tower.vision_model.encoder.layers.10.self_attn.k_proj",
93
+ "vision_tower.vision_model.encoder.layers.11.self_attn.k_proj",
94
+ "vision_tower.vision_model.encoder.layers.19.mlp.fc1",
95
+ "vision_tower.vision_model.encoder.layers.21.self_attn.v_proj",
96
+ "vision_tower.vision_model.encoder.layers.21.mlp.fc2",
97
+ "multi_modal_projector.linear_1.base_layer",
98
+ "vision_tower.vision_model.encoder.layers.1.self_attn.v_proj",
99
+ "vision_tower.vision_model.encoder.layers.1.self_attn.out_proj",
100
+ "vision_tower.vision_model.encoder.layers.9.self_attn.out_proj",
101
+ "vision_tower.vision_model.encoder.layers.14.self_attn.k_proj",
102
+ "vision_tower.vision_model.encoder.layers.18.self_attn.out_proj",
103
+ "vision_tower.vision_model.encoder.layers.19.self_attn.out_proj",
104
+ "vision_tower.vision_model.encoder.layers.10.self_attn.out_proj",
105
+ "vision_tower.vision_model.encoder.layers.1.self_attn.q_proj",
106
+ "multi_modal_projector.linear_2.lora_B.default",
107
+ "vision_tower.vision_model.encoder.layers.23.mlp.fc2",
108
+ "vision_tower.vision_model.encoder.layers.20.mlp.fc1",
109
+ "vision_tower.vision_model.encoder.layers.23.self_attn.k_proj",
110
+ "vision_tower.vision_model.encoder.layers.2.self_attn.k_proj",
111
+ "vision_tower.vision_model.encoder.layers.5.self_attn.v_proj",
112
+ "vision_tower.vision_model.encoder.layers.17.self_attn.v_proj",
113
+ "vision_tower.vision_model.encoder.layers.12.mlp.fc2",
114
+ "vision_tower.vision_model.encoder.layers.7.mlp.fc2",
115
+ "vision_tower.vision_model.encoder.layers.3.self_attn.v_proj",
116
+ "vision_tower.vision_model.encoder.layers.16.mlp.fc2",
117
+ "vision_tower.vision_model.encoder.layers.23.self_attn.out_proj",
118
+ "multi_modal_projector.linear_1.lora_B.default",
119
+ "vision_tower.vision_model.encoder.layers.21.self_attn.out_proj",
120
+ "vision_tower.vision_model.encoder.layers.10.mlp.fc2",
121
+ "vision_tower.vision_model.encoder.layers.2.mlp.fc1",
122
+ "vision_tower.vision_model.encoder.layers.0.self_attn.q_proj",
123
+ "vision_tower.vision_model.encoder.layers.9.self_attn.k_proj",
124
+ "vision_tower.vision_model.encoder.layers.15.self_attn.k_proj",
125
+ "vision_tower.vision_model.encoder.layers.23.self_attn.q_proj",
126
+ "vision_tower.vision_model.encoder.layers.7.self_attn.out_proj",
127
+ "vision_tower.vision_model.encoder.layers.14.mlp.fc2",
128
+ "vision_tower.vision_model.encoder.layers.11.self_attn.q_proj",
129
+ "vision_tower.vision_model.encoder.layers.6.self_attn.q_proj",
130
+ "multi_modal_projector.linear_2.lora_A.default",
131
+ "vision_tower.vision_model.encoder.layers.16.self_attn.q_proj",
132
+ "vision_tower.vision_model.encoder.layers.11.self_attn.out_proj",
133
+ "vision_tower.vision_model.encoder.layers.0.self_attn.out_proj",
134
+ "vision_tower.vision_model.encoder.layers.15.mlp.fc1",
135
+ "vision_tower.vision_model.encoder.layers.18.self_attn.k_proj",
136
+ "vision_tower.vision_model.encoder.layers.22.mlp.fc1",
137
+ "vision_tower.vision_model.encoder.layers.9.mlp.fc1",
138
+ "vision_tower.vision_model.encoder.layers.10.self_attn.q_proj",
139
+ "vision_tower.vision_model.encoder.layers.18.self_attn.v_proj",
140
+ "vision_tower.vision_model.encoder.layers.18.mlp.fc1",
141
+ "vision_tower.vision_model.encoder.layers.4.self_attn.out_proj",
142
+ "vision_tower.vision_model.encoder.layers.13.self_attn.k_proj",
143
+ "vision_tower.vision_model.encoder.layers.4.self_attn.q_proj",
144
+ "vision_tower.vision_model.encoder.layers.7.self_attn.k_proj",
145
+ "vision_tower.vision_model.encoder.layers.17.mlp.fc1",
146
+ "vision_tower.vision_model.encoder.layers.8.self_attn.k_proj",
147
+ "vision_tower.vision_model.encoder.layers.6.self_attn.v_proj",
148
+ "vision_tower.vision_model.encoder.layers.11.mlp.fc2",
149
+ "vision_tower.vision_model.encoder.layers.12.mlp.fc1",
150
+ "vision_tower.vision_model.encoder.layers.10.mlp.fc1",
151
+ "vision_tower.vision_model.encoder.layers.6.self_attn.k_proj",
152
+ "vision_tower.vision_model.encoder.layers.16.self_attn.k_proj",
153
+ "vision_tower.vision_model.encoder.layers.2.mlp.fc2",
154
+ "vision_tower.vision_model.encoder.layers.1.mlp.fc1",
155
+ "vision_tower.vision_model.encoder.layers.8.self_attn.out_proj",
156
+ "vision_tower.vision_model.encoder.layers.16.self_attn.out_proj",
157
+ "vision_tower.vision_model.encoder.layers.22.mlp.fc2",
158
+ "vision_tower.vision_model.encoder.layers.1.self_attn.k_proj",
159
+ "vision_tower.vision_model.encoder.layers.13.self_attn.out_proj",
160
+ "vision_tower.vision_model.encoder.layers.22.self_attn.q_proj",
161
+ "vision_tower.vision_model.encoder.layers.11.self_attn.v_proj",
162
+ "vision_tower.vision_model.encoder.layers.5.self_attn.out_proj",
163
+ "vision_tower.vision_model.encoder.layers.3.self_attn.out_proj",
164
+ "vision_tower.vision_model.encoder.layers.4.self_attn.k_proj",
165
+ "vision_tower.vision_model.encoder.layers.15.self_attn.v_proj",
166
+ "vision_tower.vision_model.encoder.layers.14.mlp.fc1",
167
+ "vision_tower.vision_model.encoder.layers.2.self_attn.q_proj",
168
+ "vision_tower.vision_model.encoder.layers.17.self_attn.q_proj",
169
+ "vision_tower.vision_model.encoder.layers.12.self_attn.q_proj",
170
+ "vision_tower.vision_model.encoder.layers.22.self_attn.k_proj",
171
+ "vision_tower.vision_model.encoder.layers.5.mlp.fc2",
172
+ "vision_tower.vision_model.encoder.layers.0.mlp.fc1",
173
+ "multi_modal_projector.linear_2.base_layer",
174
+ "vision_tower.vision_model.encoder.layers.20.self_attn.q_proj",
175
+ "vision_tower.vision_model.encoder.layers.20.self_attn.out_proj"
176
+ ],
177
+ "task_type": null,
178
+ "use_dora": false,
179
+ "use_rslora": false
180
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9bd9814b91ad88ff1a79a332aea29a92cb8ad820b55c85aeea085c46ac98e943
3
+ size 1589496912
special_tokens_map.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<s>",
4
+ "lstrip": false,
5
+ "normalized": false,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "</s>",
11
+ "lstrip": false,
12
+ "normalized": false,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "pad_token": {
17
+ "content": "<pad>",
18
+ "lstrip": false,
19
+ "normalized": false,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ },
23
+ "unk_token": {
24
+ "content": "<unk>",
25
+ "lstrip": false,
26
+ "normalized": false,
27
+ "rstrip": false,
28
+ "single_word": false
29
+ }
30
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,68 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "add_prefix_space": null,
5
+ "added_tokens_decoder": {
6
+ "0": {
7
+ "content": "<unk>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false,
12
+ "special": true
13
+ },
14
+ "1": {
15
+ "content": "<s>",
16
+ "lstrip": false,
17
+ "normalized": false,
18
+ "rstrip": false,
19
+ "single_word": false,
20
+ "special": true
21
+ },
22
+ "2": {
23
+ "content": "</s>",
24
+ "lstrip": false,
25
+ "normalized": false,
26
+ "rstrip": false,
27
+ "single_word": false,
28
+ "special": true
29
+ },
30
+ "32000": {
31
+ "content": "<image>",
32
+ "lstrip": false,
33
+ "normalized": false,
34
+ "rstrip": false,
35
+ "single_word": false,
36
+ "special": true
37
+ },
38
+ "32001": {
39
+ "content": "<pad>",
40
+ "lstrip": false,
41
+ "normalized": false,
42
+ "rstrip": false,
43
+ "single_word": false,
44
+ "special": true
45
+ }
46
+ },
47
+ "additional_special_tokens": [],
48
+ "bos_token": "<s>",
49
+ "chat_template": "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}{{ raise_exception('Conversation roles must alternate user/assistant/user/assistant/...') }}{% endif %}{% if message['role'] == 'user' %}{{ '[INST] ' + message['content'] + ' [/INST]' }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token}}{% else %}{{ raise_exception('Only user and assistant roles are supported!') }}{% endif %}{% endfor %}",
50
+ "clean_up_tokenization_spaces": false,
51
+ "eos_token": "</s>",
52
+ "legacy": true,
53
+ "max_length": null,
54
+ "model_max_length": 1000000000000000019884624838656,
55
+ "pad_to_multiple_of": null,
56
+ "pad_token": "<pad>",
57
+ "pad_token_type_id": 0,
58
+ "padding_side": "right",
59
+ "processor_class": "LlavaNextProcessor",
60
+ "sp_model_kwargs": {},
61
+ "spaces_between_special_tokens": false,
62
+ "stride": 0,
63
+ "tokenizer_class": "LlamaTokenizer",
64
+ "truncation_side": "right",
65
+ "truncation_strategy": "longest_first",
66
+ "unk_token": "<unk>",
67
+ "use_default_system_prompt": false
68
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d52bfa99adfc8e14db8dae4b8f84c3ce5b4f50170d945cf980d78cbfbbcb36fe
3
+ size 5560