Hugging Face's logo

anantg
/
Mixtral-Finetune-Output

PEFT
Safetensors
trl
sft
Generated from Trainer
Model card Files Community
anantg commited on
Commit
b2b685c
•
1 Parent(s): 749cb8e

anantg/Mixtral-Finetuned

Browse files
Files changed (7) hide show
  1. README.md +168 -0
  2. adapter_config.json +28 -0
  3. adapter_model.safetensors +3 -0
  4. special_tokens_map.json +24 -0
  5. tokenizer.json +0 -0
  6. tokenizer_config.json +43 -0
  7. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,168 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ datasets:
9
+ - generator
10
+ base_model: mistralai/Mixtral-8x7B-Instruct-v0.1
11
+ model-index:
12
+ - name: Mixtral-Finetune-Output
13
+ results: []
14
+ ---
15
+
16
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
+ should probably proofread and complete it, then remove this comment. -->
18
+
19
+ # Mixtral-Finetune-Output
20
+
21
+ This model is a fine-tuned version of [mistralai/Mixtral-8x7B-Instruct-v0.1](https://huggingface.co/mistralai/Mixtral-8x7B-Instruct-v0.1) on the generator dataset.
22
+ It achieves the following results on the evaluation set:
23
+ - Loss: 1.3062
24
+
25
+ ## Model description
26
+
27
+ More information needed
28
+
29
+ ## Intended uses & limitations
30
+
31
+ More information needed
32
+
33
+ ## Training and evaluation data
34
+
35
+ More information needed
36
+
37
+ ## Training procedure
38
+
39
+ ### Training hyperparameters
40
+
41
+ The following hyperparameters were used during training:
42
+ - learning_rate: 0.0001
43
+ - train_batch_size: 1
44
+ - eval_batch_size: 8
45
+ - seed: 42
46
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
+ - lr_scheduler_type: linear
48
+ - lr_scheduler_warmup_steps: 0.03
49
+ - num_epochs: 1
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss |
54
+ |:-------------:|:-----:|:----:|:---------------:|
55
+ | 1.5745 | 0.01 | 10 | 1.4701 |
56
+ | 1.5372 | 0.02 | 20 | 1.4541 |
57
+ | 1.4147 | 0.03 | 30 | 1.4433 |
58
+ | 1.423 | 0.04 | 40 | 1.4366 |
59
+ | 1.5318 | 0.05 | 50 | 1.4326 |
60
+ | 1.334 | 0.06 | 60 | 1.4296 |
61
+ | 1.364 | 0.07 | 70 | 1.4244 |
62
+ | 1.332 | 0.08 | 80 | 1.4194 |
63
+ | 1.3742 | 0.09 | 90 | 1.4163 |
64
+ | 1.4497 | 0.1 | 100 | 1.4124 |
65
+ | 1.4145 | 0.1 | 110 | 1.4098 |
66
+ | 1.4224 | 0.11 | 120 | 1.4050 |
67
+ | 1.4013 | 0.12 | 130 | 1.4017 |
68
+ | 1.547 | 0.13 | 140 | 1.4020 |
69
+ | 1.4969 | 0.14 | 150 | 1.3967 |
70
+ | 1.5716 | 0.15 | 160 | 1.3943 |
71
+ | 1.3677 | 0.16 | 170 | 1.3915 |
72
+ | 1.3789 | 0.17 | 180 | 1.3901 |
73
+ | 1.3188 | 0.18 | 190 | 1.3869 |
74
+ | 1.3317 | 0.19 | 200 | 1.3846 |
75
+ | 1.2552 | 0.2 | 210 | 1.3809 |
76
+ | 1.2584 | 0.21 | 220 | 1.3788 |
77
+ | 1.3958 | 0.22 | 230 | 1.3776 |
78
+ | 1.3345 | 0.23 | 240 | 1.3755 |
79
+ | 1.3562 | 0.24 | 250 | 1.3723 |
80
+ | 1.343 | 0.25 | 260 | 1.3726 |
81
+ | 1.3705 | 0.26 | 270 | 1.3695 |
82
+ | 1.5719 | 0.27 | 280 | 1.3687 |
83
+ | 1.3634 | 0.28 | 290 | 1.3652 |
84
+ | 1.4465 | 0.29 | 300 | 1.3668 |
85
+ | 1.3949 | 0.29 | 310 | 1.3642 |
86
+ | 1.3147 | 0.3 | 320 | 1.3631 |
87
+ | 1.368 | 0.31 | 330 | 1.3613 |
88
+ | 1.3482 | 0.32 | 340 | 1.3603 |
89
+ | 1.3143 | 0.33 | 350 | 1.3591 |
90
+ | 1.4717 | 0.34 | 360 | 1.3568 |
91
+ | 1.2089 | 0.35 | 370 | 1.3555 |
92
+ | 1.4223 | 0.36 | 380 | 1.3529 |
93
+ | 1.3895 | 0.37 | 390 | 1.3523 |
94
+ | 1.309 | 0.38 | 400 | 1.3504 |
95
+ | 1.3698 | 0.39 | 410 | 1.3487 |
96
+ | 1.2834 | 0.4 | 420 | 1.3468 |
97
+ | 1.2747 | 0.41 | 430 | 1.3471 |
98
+ | 1.3167 | 0.42 | 440 | 1.3460 |
99
+ | 1.3232 | 0.43 | 450 | 1.3438 |
100
+ | 1.3628 | 0.44 | 460 | 1.3422 |
101
+ | 1.3828 | 0.45 | 470 | 1.3417 |
102
+ | 1.3756 | 0.46 | 480 | 1.3412 |
103
+ | 1.385 | 0.47 | 490 | 1.3418 |
104
+ | 1.3622 | 0.48 | 500 | 1.3392 |
105
+ | 1.3322 | 0.49 | 510 | 1.3381 |
106
+ | 1.368 | 0.49 | 520 | 1.3365 |
107
+ | 1.3373 | 0.5 | 530 | 1.3355 |
108
+ | 1.4931 | 0.51 | 540 | 1.3354 |
109
+ | 1.3986 | 0.52 | 550 | 1.3333 |
110
+ | 1.3053 | 0.53 | 560 | 1.3312 |
111
+ | 1.2736 | 0.54 | 570 | 1.3297 |
112
+ | 1.2903 | 0.55 | 580 | 1.3298 |
113
+ | 1.328 | 0.56 | 590 | 1.3290 |
114
+ | 1.4081 | 0.57 | 600 | 1.3290 |
115
+ | 1.2852 | 0.58 | 610 | 1.3279 |
116
+ | 1.3636 | 0.59 | 620 | 1.3268 |
117
+ | 1.3448 | 0.6 | 630 | 1.3265 |
118
+ | 1.2061 | 0.61 | 640 | 1.3252 |
119
+ | 1.3519 | 0.62 | 650 | 1.3244 |
120
+ | 1.3632 | 0.63 | 660 | 1.3248 |
121
+ | 1.3784 | 0.64 | 670 | 1.3238 |
122
+ | 1.3349 | 0.65 | 680 | 1.3216 |
123
+ | 1.2603 | 0.66 | 690 | 1.3215 |
124
+ | 1.3566 | 0.67 | 700 | 1.3224 |
125
+ | 1.316 | 0.68 | 710 | 1.3208 |
126
+ | 1.1818 | 0.69 | 720 | 1.3203 |
127
+ | 1.3631 | 0.69 | 730 | 1.3190 |
128
+ | 1.3234 | 0.7 | 740 | 1.3184 |
129
+ | 1.2759 | 0.71 | 750 | 1.3177 |
130
+ | 1.3332 | 0.72 | 760 | 1.3177 |
131
+ | 1.2764 | 0.73 | 770 | 1.3165 |
132
+ | 1.2056 | 0.74 | 780 | 1.3155 |
133
+ | 1.4285 | 0.75 | 790 | 1.3158 |
134
+ | 1.3733 | 0.76 | 800 | 1.3150 |
135
+ | 1.2735 | 0.77 | 810 | 1.3143 |
136
+ | 1.3502 | 0.78 | 820 | 1.3137 |
137
+ | 1.093 | 0.79 | 830 | 1.3130 |
138
+ | 1.3451 | 0.8 | 840 | 1.3123 |
139
+ | 1.2942 | 0.81 | 850 | 1.3119 |
140
+ | 1.3258 | 0.82 | 860 | 1.3117 |
141
+ | 1.2139 | 0.83 | 870 | 1.3114 |
142
+ | 1.2773 | 0.84 | 880 | 1.3109 |
143
+ | 1.2324 | 0.85 | 890 | 1.3101 |
144
+ | 1.4134 | 0.86 | 900 | 1.3097 |
145
+ | 1.3464 | 0.87 | 910 | 1.3095 |
146
+ | 1.2972 | 0.88 | 920 | 1.3090 |
147
+ | 1.3305 | 0.88 | 930 | 1.3086 |
148
+ | 1.3394 | 0.89 | 940 | 1.3082 |
149
+ | 1.3666 | 0.9 | 950 | 1.3078 |
150
+ | 1.3703 | 0.91 | 960 | 1.3077 |
151
+ | 1.3019 | 0.92 | 970 | 1.3077 |
152
+ | 1.2618 | 0.93 | 980 | 1.3073 |
153
+ | 1.2808 | 0.94 | 990 | 1.3071 |
154
+ | 1.2927 | 0.95 | 1000 | 1.3069 |
155
+ | 1.2688 | 0.96 | 1010 | 1.3067 |
156
+ | 1.3312 | 0.97 | 1020 | 1.3065 |
157
+ | 1.2406 | 0.98 | 1030 | 1.3064 |
158
+ | 1.3341 | 0.99 | 1040 | 1.3062 |
159
+ | 1.3531 | 1.0 | 1050 | 1.3062 |
160
+
161
+
162
+ ### Framework versions
163
+
164
+ - PEFT 0.7.1
165
+ - Transformers 4.36.2
166
+ - Pytorch 2.1.0+cu121
167
+ - Datasets 2.16.1
168
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "mistralai/Mixtral-8x7B-Instruct-v0.1",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 32,
13
+ "lora_dropout": 0.1,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "self_attn.o_proj",
23
+ "self_attn.k_proj",
24
+ "self_attn.v_proj",
25
+ "self_attn.q_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM"
28
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f38ce8cd4bda4f6cb0ae2446e04915778c4996ec00325313fe9410f85829d603
3
+ size 27297032
special_tokens_map.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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": "</s>",
17
+ "unk_token": {
18
+ "content": "<unk>",
19
+ "lstrip": false,
20
+ "normalized": false,
21
+ "rstrip": false,
22
+ "single_word": false
23
+ }
24
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ }
29
+ },
30
+ "additional_special_tokens": [],
31
+ "bos_token": "<s>",
32
+ "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 %}",
33
+ "clean_up_tokenization_spaces": false,
34
+ "eos_token": "</s>",
35
+ "legacy": true,
36
+ "model_max_length": 1000000000000000019884624838656,
37
+ "pad_token": "</s>",
38
+ "sp_model_kwargs": {},
39
+ "spaces_between_special_tokens": false,
40
+ "tokenizer_class": "LlamaTokenizer",
41
+ "unk_token": "<unk>",
42
+ "use_default_system_prompt": false
43
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c0db1c27b4e171bc8bf399197cd830184b9ebcdf54c7cc4010a111f40ee53b2e
3
+ size 4664