Hugging Face's logo

allstax
/
CodeExplainer-7b-v0.11-GPTQ

PEFT
Safetensors
Generated from Trainer
Model card Files
xet
Community
allstax commited on
Commit
c1d5794
·
1 Parent(s): 877ef83

Pushing fine-tuned model

Browse files
Files changed (4) hide show
  1. README.md +160 -0
  2. adapter_config.json +29 -0
  3. adapter_model.safetensors +3 -0
  4. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,160 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - generated_from_trainer
6
+ base_model: TheBloke/Marcoroni-7B-v3-GPTQ
7
+ model-index:
8
+ - name: 20231215-144935
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
+ # 20231215-144935
16
+
17
+ This model is a fine-tuned version of [TheBloke/Marcoroni-7B-v3-GPTQ](https://huggingface.co/TheBloke/Marcoroni-7B-v3-GPTQ) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.2792
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: 2.5e-05
39
+ - train_batch_size: 2
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
43
+ - lr_scheduler_type: linear
44
+ - lr_scheduler_warmup_steps: 2
45
+ - training_steps: 100
46
+ - mixed_precision_training: Native AMP
47
+
48
+ ### Training results
49
+
50
+ | Training Loss | Epoch | Step | Validation Loss |
51
+ |:-------------:|:-----:|:----:|:---------------:|
52
+ | 2.1951 | 0.02 | 1 | 2.0427 |
53
+ | 1.4965 | 0.03 | 2 | 2.0427 |
54
+ | 2.2716 | 0.05 | 3 | 1.9661 |
55
+ | 1.9397 | 0.06 | 4 | 1.9661 |
56
+ | 1.8659 | 0.08 | 5 | 1.9661 |
57
+ | 1.9903 | 0.09 | 6 | 1.8551 |
58
+ | 1.9031 | 0.11 | 7 | 1.7846 |
59
+ | 1.929 | 0.12 | 8 | 1.7253 |
60
+ | 0.9994 | 0.14 | 9 | 1.6749 |
61
+ | 1.4887 | 0.15 | 10 | 1.6266 |
62
+ | 1.2112 | 0.17 | 11 | 1.5763 |
63
+ | 1.3437 | 0.18 | 12 | 1.5326 |
64
+ | 1.1061 | 0.2 | 13 | 1.5014 |
65
+ | 1.4769 | 0.22 | 14 | 1.4676 |
66
+ | 1.452 | 0.23 | 15 | 1.4313 |
67
+ | 1.2393 | 0.25 | 16 | 1.3963 |
68
+ | 1.1998 | 0.26 | 17 | 1.3662 |
69
+ | 1.529 | 0.28 | 18 | 1.3353 |
70
+ | 0.949 | 0.29 | 19 | 1.3088 |
71
+ | 1.1857 | 0.31 | 20 | 1.2798 |
72
+ | 1.4852 | 0.32 | 21 | 1.2484 |
73
+ | 1.2845 | 0.34 | 22 | 1.2146 |
74
+ | 1.332 | 0.35 | 23 | 1.1795 |
75
+ | 0.8718 | 0.37 | 24 | 1.1435 |
76
+ | 1.2038 | 0.38 | 25 | 1.1069 |
77
+ | 1.2105 | 0.4 | 26 | 1.0683 |
78
+ | 0.9161 | 0.42 | 27 | 1.0277 |
79
+ | 1.0415 | 0.43 | 28 | 0.9849 |
80
+ | 1.0367 | 0.45 | 29 | 0.9400 |
81
+ | 0.9279 | 0.46 | 30 | 0.8924 |
82
+ | 0.82 | 0.48 | 31 | 0.8467 |
83
+ | 0.8286 | 0.49 | 32 | 0.8077 |
84
+ | 0.7855 | 0.51 | 33 | 0.7645 |
85
+ | 0.7919 | 0.52 | 34 | 0.7164 |
86
+ | 0.6294 | 0.54 | 35 | 0.6712 |
87
+ | 0.6205 | 0.55 | 36 | 0.6302 |
88
+ | 0.5754 | 0.57 | 37 | 0.5926 |
89
+ | 0.6331 | 0.58 | 38 | 0.5578 |
90
+ | 0.5068 | 0.6 | 39 | 0.5251 |
91
+ | 0.5252 | 0.62 | 40 | 0.5009 |
92
+ | 0.442 | 0.63 | 41 | 0.4800 |
93
+ | 0.4451 | 0.65 | 42 | 0.4628 |
94
+ | 0.3742 | 0.66 | 43 | 0.4491 |
95
+ | 0.8263 | 0.68 | 44 | 0.4383 |
96
+ | 0.4392 | 0.69 | 45 | 0.4277 |
97
+ | 0.5359 | 0.71 | 46 | 0.4174 |
98
+ | 0.3903 | 0.72 | 47 | 0.4051 |
99
+ | 0.4177 | 0.74 | 48 | 0.3921 |
100
+ | 0.4606 | 0.75 | 49 | 0.3792 |
101
+ | 0.5233 | 0.77 | 50 | 0.3700 |
102
+ | 0.3838 | 0.78 | 51 | 0.3638 |
103
+ | 0.353 | 0.8 | 52 | 0.3587 |
104
+ | 0.3108 | 0.82 | 53 | 0.3530 |
105
+ | 0.4029 | 0.83 | 54 | 0.3470 |
106
+ | 0.3425 | 0.85 | 55 | 0.3411 |
107
+ | 0.2976 | 0.86 | 56 | 0.3345 |
108
+ | 0.5347 | 0.88 | 57 | 0.3291 |
109
+ | 0.3208 | 0.89 | 58 | 0.3239 |
110
+ | 0.3319 | 0.91 | 59 | 0.3194 |
111
+ | 0.3333 | 0.92 | 60 | 0.3160 |
112
+ | 0.3102 | 0.94 | 61 | 0.3135 |
113
+ | 0.4149 | 0.95 | 62 | 0.3118 |
114
+ | 0.3044 | 0.97 | 63 | 0.3098 |
115
+ | 0.2688 | 0.98 | 64 | 0.3076 |
116
+ | 0.3926 | 1.0 | 65 | 0.3060 |
117
+ | 0.2969 | 1.02 | 66 | 0.3045 |
118
+ | 0.3799 | 1.03 | 67 | 0.3029 |
119
+ | 0.3045 | 1.05 | 68 | 0.3014 |
120
+ | 0.2295 | 1.06 | 69 | 0.2997 |
121
+ | 0.4113 | 1.08 | 70 | 0.2983 |
122
+ | 0.3093 | 1.09 | 71 | 0.2969 |
123
+ | 0.2542 | 1.11 | 72 | 0.2957 |
124
+ | 0.2445 | 1.12 | 73 | 0.2947 |
125
+ | 0.2592 | 1.14 | 74 | 0.2935 |
126
+ | 0.6259 | 1.15 | 75 | 0.2925 |
127
+ | 0.3217 | 1.17 | 76 | 0.2916 |
128
+ | 0.2969 | 1.18 | 77 | 0.2907 |
129
+ | 0.287 | 1.2 | 78 | 0.2898 |
130
+ | 0.2642 | 1.22 | 79 | 0.2890 |
131
+ | 0.2735 | 1.23 | 80 | 0.2880 |
132
+ | 0.2472 | 1.25 | 81 | 0.2870 |
133
+ | 0.3068 | 1.26 | 82 | 0.2862 |
134
+ | 0.2779 | 1.28 | 83 | 0.2855 |
135
+ | 0.2915 | 1.29 | 84 | 0.2848 |
136
+ | 0.3926 | 1.31 | 85 | 0.2841 |
137
+ | 0.3057 | 1.32 | 86 | 0.2836 |
138
+ | 0.2413 | 1.34 | 87 | 0.2830 |
139
+ | 0.2851 | 1.35 | 88 | 0.2826 |
140
+ | 0.2295 | 1.37 | 89 | 0.2821 |
141
+ | 0.2611 | 1.38 | 90 | 0.2817 |
142
+ | 0.6346 | 1.4 | 91 | 0.2814 |
143
+ | 0.2297 | 1.42 | 92 | 0.2811 |
144
+ | 0.4868 | 1.43 | 93 | 0.2808 |
145
+ | 0.2819 | 1.45 | 94 | 0.2805 |
146
+ | 0.2589 | 1.46 | 95 | 0.2803 |
147
+ | 0.2171 | 1.48 | 96 | 0.2801 |
148
+ | 0.2581 | 1.49 | 97 | 0.2798 |
149
+ | 0.2495 | 1.51 | 98 | 0.2796 |
150
+ | 0.2702 | 1.52 | 99 | 0.2793 |
151
+ | 0.271 | 1.54 | 100 | 0.2792 |
152
+
153
+
154
+ ### Framework versions
155
+
156
+ - PEFT 0.7.2.dev0
157
+ - Transformers 4.37.0.dev0
158
+ - Pytorch 2.1.0+cu121
159
+ - Datasets 2.15.0
160
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "TheBloke/Marcoroni-7B-v3-GPTQ",
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": 8,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "k_proj",
23
+ "v_proj",
24
+ "o_proj",
25
+ "q_proj"
26
+ ],
27
+ "task_type": "CAUSAL_LM",
28
+ "use_rslora": false
29
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d64ff06fd94f533ec80293c1a1b75c2f43b8fb49d5cea2c125a87d08ecf68370
3
+ size 27297032
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:02e6b3862d60150bebb47b8a45548ff1d0f24795109f1147f66259a6fc3212ee
3
+ size 4728