Hugging Face's logo

Gunslinger3D
/
fine-tuning-dolphin-mistral-with-webglm-qa-with-lora_1

PEFT
Safetensors
Generated from Trainer
Model card Files Community
Gunslinger3D commited on
Commit
4880144
·
verified ·
1 Parent(s): a3e5bc7

fine-tuning-dolphin-mistral-with-webglm-qa-with-lora_1

Browse files
Files changed (4) hide show
  1. README.md +132 -0
  2. adapter_config.json +30 -0
  3. adapter_model.safetensors +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,132 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: cognitivecomputations/dolphin-2.8-mistral-7b-v02
7
+ model-index:
8
+ - name: fine-tuning-dolphin-mistral-with-webglm-qa-with-lora_1
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # fine-tuning-dolphin-mistral-with-webglm-qa-with-lora_1
16
+
17
+ This model is a fine-tuned version of [cognitivecomputations/dolphin-2.8-mistral-7b-v02](https://huggingface.co/cognitivecomputations/dolphin-2.8-mistral-7b-v02) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.2999
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 2
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 5
43
+ - total_train_batch_size: 10
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: linear
46
+ - lr_scheduler_warmup_steps: 60
47
+ - training_steps: 700
48
+ - mixed_precision_training: Native AMP
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss |
53
+ |:-------------:|:-----:|:----:|:---------------:|
54
+ | 1.7558 | 0.16 | 10 | 1.4842 |
55
+ | 1.4966 | 0.32 | 20 | 1.3367 |
56
+ | 1.2328 | 0.48 | 30 | 1.1282 |
57
+ | 0.9873 | 0.64 | 40 | 1.0817 |
58
+ | 0.9661 | 0.8 | 50 | 0.9967 |
59
+ | 0.8808 | 0.96 | 60 | 0.8844 |
60
+ | 0.7455 | 1.13 | 70 | 0.7337 |
61
+ | 0.6018 | 1.29 | 80 | 0.6164 |
62
+ | 0.4899 | 1.45 | 90 | 0.5440 |
63
+ | 0.4402 | 1.61 | 100 | 0.4971 |
64
+ | 0.4154 | 1.77 | 110 | 0.4555 |
65
+ | 0.4025 | 1.93 | 120 | 0.4238 |
66
+ | 0.3992 | 2.09 | 130 | 0.4007 |
67
+ | 0.3585 | 2.25 | 140 | 0.3862 |
68
+ | 0.3369 | 2.41 | 150 | 0.3666 |
69
+ | 0.3328 | 2.57 | 160 | 0.3537 |
70
+ | 0.3216 | 2.73 | 170 | 0.3423 |
71
+ | 0.2859 | 2.89 | 180 | 0.3303 |
72
+ | 0.2967 | 3.05 | 190 | 0.3211 |
73
+ | 0.2933 | 3.22 | 200 | 0.3114 |
74
+ | 0.2716 | 3.38 | 210 | 0.3097 |
75
+ | 0.255 | 3.54 | 220 | 0.3053 |
76
+ | 0.2731 | 3.7 | 230 | 0.2990 |
77
+ | 0.2729 | 3.86 | 240 | 0.2972 |
78
+ | 0.2701 | 4.02 | 250 | 0.3030 |
79
+ | 0.2558 | 4.18 | 260 | 0.3042 |
80
+ | 0.2612 | 4.34 | 270 | 0.3301 |
81
+ | 0.3048 | 4.5 | 280 | 0.4564 |
82
+ | 0.5437 | 4.66 | 290 | 0.7938 |
83
+ | 1.5888 | 4.82 | 300 | 1.5418 |
84
+ | 0.6588 | 4.98 | 310 | 0.4630 |
85
+ | 0.5345 | 5.14 | 320 | 0.9088 |
86
+ | 1.1475 | 5.31 | 330 | 1.6381 |
87
+ | 1.6442 | 5.47 | 340 | 2.0495 |
88
+ | 2.2517 | 5.63 | 350 | 1.7558 |
89
+ | 0.9492 | 5.79 | 360 | 0.5187 |
90
+ | 0.3727 | 5.95 | 370 | 0.3763 |
91
+ | 0.3139 | 6.11 | 380 | 0.3376 |
92
+ | 0.2896 | 6.27 | 390 | 0.3195 |
93
+ | 0.283 | 6.43 | 400 | 0.3106 |
94
+ | 0.2646 | 6.59 | 410 | 0.3105 |
95
+ | 0.2674 | 6.75 | 420 | 0.3256 |
96
+ | 0.3482 | 6.91 | 430 | 0.4016 |
97
+ | 0.4193 | 7.07 | 440 | 0.6300 |
98
+ | 0.7397 | 7.23 | 450 | 1.0617 |
99
+ | 1.1954 | 7.4 | 460 | 1.6157 |
100
+ | 1.6177 | 7.56 | 470 | 1.8019 |
101
+ | 1.2996 | 7.72 | 480 | 0.9151 |
102
+ | 0.6605 | 7.88 | 490 | 0.5433 |
103
+ | 0.416 | 8.04 | 500 | 0.4012 |
104
+ | 0.3412 | 8.2 | 510 | 0.3685 |
105
+ | 0.3322 | 8.36 | 520 | 0.3928 |
106
+ | 0.3516 | 8.52 | 530 | 0.3641 |
107
+ | 0.3406 | 8.68 | 540 | 0.4061 |
108
+ | 0.3772 | 8.84 | 550 | 0.4145 |
109
+ | 0.3695 | 9.0 | 560 | 0.5453 |
110
+ | 0.5824 | 9.16 | 570 | 0.7332 |
111
+ | 0.5139 | 9.32 | 580 | 0.4839 |
112
+ | 0.3798 | 9.49 | 590 | 0.3758 |
113
+ | 0.319 | 9.65 | 600 | 0.3438 |
114
+ | 0.3082 | 9.81 | 610 | 0.3301 |
115
+ | 0.3017 | 9.97 | 620 | 0.3225 |
116
+ | 0.2862 | 10.13 | 630 | 0.3156 |
117
+ | 0.2586 | 10.29 | 640 | 0.3109 |
118
+ | 0.2878 | 10.45 | 650 | 0.3082 |
119
+ | 0.2766 | 10.61 | 660 | 0.3056 |
120
+ | 0.2834 | 10.77 | 670 | 0.3042 |
121
+ | 0.2513 | 10.93 | 680 | 0.3020 |
122
+ | 0.2762 | 11.09 | 690 | 0.3007 |
123
+ | 0.28 | 11.25 | 700 | 0.2999 |
124
+
125
+
126
+ ### Framework versions
127
+
128
+ - PEFT 0.7.1
129
+ - Transformers 4.36.2
130
+ - Pytorch 2.0.0
131
+ - Datasets 2.15.0
132
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,30 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "cognitivecomputations/dolphin-2.8-mistral-7b-v02",
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": 0.05,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 16,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "q_proj",
23
+ "fc2",
24
+ "dense",
25
+ "v_proj",
26
+ "fc1",
27
+ "k_proj"
28
+ ],
29
+ "task_type": "CAUSAL_LM"
30
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f087974e5e7e9535f960953d97c9d6bae4861946f6917f9a2b406466554c99dd
3
+ size 37774528
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e7f63bda251e2401a5b628f036d756a625509055d9bdf0b871018ca16967473
3
+ size 4347