Hugging Face's logo

Weni
/
ZeroShot-3.3.2-Mistral-7b-Multilanguage-3.2.0

PEFT
Safetensors
mistral
trl
sft
Generated from Trainer
Model card Files
xet
 Community
ironrock commited on
Commit
ebfcd74
·
verified ·
1 Parent(s): a693d78

Model save

Browse files
Files changed (1) hide show
  1. README.md +160 -0
README.md ADDED
@@ -0,0 +1,160 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ library_name: peft
4
+ tags:
5
+ - trl
6
+ - sft
7
+ - generated_from_trainer
8
+ base_model: mistralai/Mistral-7B-Instruct-v0.2
9
+ model-index:
10
+ - name: ZeroShot-3.3.2-Mistral-7b-Multilanguage-3.1.0
11
+ results: []
12
+ ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # ZeroShot-3.3.2-Mistral-7b-Multilanguage-3.1.0
18
+
19
+ This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on an unknown dataset.
20
+ It achieves the following results on the evaluation set:
21
+ - Loss: 0.3386
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0002
41
+ - train_batch_size: 8
42
+ - eval_batch_size: 2
43
+ - seed: 42
44
+ - gradient_accumulation_steps: 2
45
+ - total_train_batch_size: 16
46
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
+ - lr_scheduler_type: cosine
48
+ - lr_scheduler_warmup_ratio: 0.1
49
+ - num_epochs: 3
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Validation Loss |
55
+ |:-------------:|:-----:|:----:|:---------------:|
56
+ | 1.9008 | 0.03 | 50 | 1.3557 |
57
+ | 0.724 | 0.06 | 100 | 0.5604 |
58
+ | 0.5279 | 0.09 | 150 | 0.5183 |
59
+ | 0.4864 | 0.12 | 200 | 0.4832 |
60
+ | 0.4598 | 0.16 | 250 | 0.4487 |
61
+ | 0.4286 | 0.19 | 300 | 0.4403 |
62
+ | 0.4463 | 0.22 | 350 | 0.4362 |
63
+ | 0.4279 | 0.25 | 400 | 0.4321 |
64
+ | 0.4252 | 0.28 | 450 | 0.4273 |
65
+ | 0.4214 | 0.31 | 500 | 0.4246 |
66
+ | 0.4198 | 0.34 | 550 | 0.4209 |
67
+ | 0.4152 | 0.37 | 600 | 0.4169 |
68
+ | 0.4114 | 0.4 | 650 | 0.4138 |
69
+ | 0.4197 | 0.43 | 700 | 0.4099 |
70
+ | 0.4102 | 0.47 | 750 | 0.4081 |
71
+ | 0.3914 | 0.5 | 800 | 0.4052 |
72
+ | 0.4038 | 0.53 | 850 | 0.4025 |
73
+ | 0.3941 | 0.56 | 900 | 0.4011 |
74
+ | 0.3989 | 0.59 | 950 | 0.3990 |
75
+ | 0.3947 | 0.62 | 1000 | 0.3968 |
76
+ | 0.3903 | 0.65 | 1050 | 0.3954 |
77
+ | 0.3903 | 0.68 | 1100 | 0.3931 |
78
+ | 0.3881 | 0.71 | 1150 | 0.3922 |
79
+ | 0.3928 | 0.74 | 1200 | 0.3901 |
80
+ | 0.3769 | 0.78 | 1250 | 0.3880 |
81
+ | 0.3717 | 0.81 | 1300 | 0.3860 |
82
+ | 0.3697 | 0.84 | 1350 | 0.3851 |
83
+ | 0.3666 | 0.87 | 1400 | 0.3834 |
84
+ | 0.3834 | 0.9 | 1450 | 0.3815 |
85
+ | 0.3777 | 0.93 | 1500 | 0.3801 |
86
+ | 0.3678 | 0.96 | 1550 | 0.3779 |
87
+ | 0.3779 | 0.99 | 1600 | 0.3777 |
88
+ | 0.3547 | 1.02 | 1650 | 0.3764 |
89
+ | 0.3463 | 1.05 | 1700 | 0.3749 |
90
+ | 0.3386 | 1.09 | 1750 | 0.3739 |
91
+ | 0.3493 | 1.12 | 1800 | 0.3737 |
92
+ | 0.3527 | 1.15 | 1850 | 0.3717 |
93
+ | 0.3471 | 1.18 | 1900 | 0.3712 |
94
+ | 0.3414 | 1.21 | 1950 | 0.3704 |
95
+ | 0.3464 | 1.24 | 2000 | 0.3683 |
96
+ | 0.3379 | 1.27 | 2050 | 0.3682 |
97
+ | 0.3469 | 1.3 | 2100 | 0.3665 |
98
+ | 0.3311 | 1.33 | 2150 | 0.3659 |
99
+ | 0.3377 | 1.36 | 2200 | 0.3644 |
100
+ | 0.3375 | 1.4 | 2250 | 0.3629 |
101
+ | 0.3415 | 1.43 | 2300 | 0.3619 |
102
+ | 0.3429 | 1.46 | 2350 | 0.3607 |
103
+ | 0.3316 | 1.49 | 2400 | 0.3607 |
104
+ | 0.3339 | 1.52 | 2450 | 0.3588 |
105
+ | 0.3438 | 1.55 | 2500 | 0.3581 |
106
+ | 0.3403 | 1.58 | 2550 | 0.3572 |
107
+ | 0.3343 | 1.61 | 2600 | 0.3555 |
108
+ | 0.3396 | 1.64 | 2650 | 0.3545 |
109
+ | 0.3349 | 1.67 | 2700 | 0.3537 |
110
+ | 0.3285 | 1.71 | 2750 | 0.3527 |
111
+ | 0.3241 | 1.74 | 2800 | 0.3518 |
112
+ | 0.3306 | 1.77 | 2850 | 0.3512 |
113
+ | 0.3265 | 1.8 | 2900 | 0.3499 |
114
+ | 0.3276 | 1.83 | 2950 | 0.3491 |
115
+ | 0.3259 | 1.86 | 3000 | 0.3486 |
116
+ | 0.3281 | 1.89 | 3050 | 0.3477 |
117
+ | 0.3199 | 1.92 | 3100 | 0.3470 |
118
+ | 0.3315 | 1.95 | 3150 | 0.3457 |
119
+ | 0.3306 | 1.98 | 3200 | 0.3455 |
120
+ | 0.306 | 2.02 | 3250 | 0.3463 |
121
+ | 0.2975 | 2.05 | 3300 | 0.3455 |
122
+ | 0.2906 | 2.08 | 3350 | 0.3457 |
123
+ | 0.2942 | 2.11 | 3400 | 0.3454 |
124
+ | 0.2898 | 2.14 | 3450 | 0.3450 |
125
+ | 0.299 | 2.17 | 3500 | 0.3446 |
126
+ | 0.2913 | 2.2 | 3550 | 0.3436 |
127
+ | 0.2891 | 2.23 | 3600 | 0.3429 |
128
+ | 0.2875 | 2.26 | 3650 | 0.3439 |
129
+ | 0.2838 | 2.29 | 3700 | 0.3426 |
130
+ | 0.2944 | 2.33 | 3750 | 0.3424 |
131
+ | 0.2904 | 2.36 | 3800 | 0.3424 |
132
+ | 0.2926 | 2.39 | 3850 | 0.3420 |
133
+ | 0.2992 | 2.42 | 3900 | 0.3413 |
134
+ | 0.2834 | 2.45 | 3950 | 0.3412 |
135
+ | 0.2923 | 2.48 | 4000 | 0.3406 |
136
+ | 0.291 | 2.51 | 4050 | 0.3401 |
137
+ | 0.2868 | 2.54 | 4100 | 0.3402 |
138
+ | 0.2867 | 2.57 | 4150 | 0.3398 |
139
+ | 0.2837 | 2.6 | 4200 | 0.3399 |
140
+ | 0.288 | 2.64 | 4250 | 0.3393 |
141
+ | 0.2874 | 2.67 | 4300 | 0.3393 |
142
+ | 0.2866 | 2.7 | 4350 | 0.3392 |
143
+ | 0.2884 | 2.73 | 4400 | 0.3390 |
144
+ | 0.2862 | 2.76 | 4450 | 0.3389 |
145
+ | 0.2938 | 2.79 | 4500 | 0.3389 |
146
+ | 0.3009 | 2.82 | 4550 | 0.3387 |
147
+ | 0.2896 | 2.85 | 4600 | 0.3387 |
148
+ | 0.2902 | 2.88 | 4650 | 0.3386 |
149
+ | 0.2891 | 2.91 | 4700 | 0.3386 |
150
+ | 0.2926 | 2.95 | 4750 | 0.3386 |
151
+ | 0.2868 | 2.98 | 4800 | 0.3386 |
152
+
153
+
154
+ ### Framework versions
155
+
156
+ - PEFT 0.8.2
157
+ - Transformers 4.38.0
158
+ - Pytorch 2.1.0+cu121
159
+ - Datasets 2.17.1
160
+ - Tokenizers 0.15.2