Hugging Face's logo

thrunlab
/
Mistral_Sparse_refined_web_50p_graceful_True

Text Generation
Transformers
Safetensors
sparse_mistral
Generated from Trainer
custom_code
Model card Files Community
lukeleeai commited on
Commit
c7c20c6
·
verified ·
1 Parent(s): e268898

End of training

Browse files
Files changed (4) hide show
  1. README.md +52 -202
  2. model-00001-of-00003.safetensors +1 -1
  3. model-00002-of-00003.safetensors +1 -1
  4. model-00003-of-00003.safetensors +1 -1
README.md CHANGED
@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
- - Loss: 2.3270
19
 
20
  ## Model description
21
 
@@ -45,212 +45,62 @@ The following hyperparameters were used during training:
45
  - total_eval_batch_size: 4
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
- - training_steps: 5000
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss |
53
  |:-------------:|:-----:|:----:|:---------------:|
54
- | 3.7944 | 0.0 | 25 | 2.4024 |
55
- | 3.7251 | 0.01 | 50 | 2.3892 |
56
- | 2.3685 | 0.01 | 75 | 2.5422 |
57
- | 2.2521 | 0.02 | 100 | 2.4731 |
58
- | 2.2468 | 0.02 | 125 | 2.4459 |
59
- | 2.3204 | 0.02 | 150 | 2.4365 |
60
- | 2.2561 | 0.03 | 175 | 2.4258 |
61
- | 2.2388 | 0.03 | 200 | 2.4174 |
62
- | 2.2738 | 0.04 | 225 | 2.4128 |
63
- | 2.3089 | 0.04 | 250 | 2.4093 |
64
- | 2.2252 | 0.04 | 275 | 2.4057 |
65
- | 2.2394 | 0.05 | 300 | 2.3994 |
66
- | 2.2531 | 0.05 | 325 | 2.4003 |
67
- | 2.0899 | 0.06 | 350 | 2.4004 |
68
- | 2.2479 | 0.06 | 375 | 2.3982 |
69
- | 2.2875 | 0.06 | 400 | 2.3987 |
70
- | 2.282 | 0.07 | 425 | 2.3959 |
71
- | 2.2434 | 0.07 | 450 | 2.3920 |
72
- | 2.1592 | 0.08 | 475 | 2.3930 |
73
- | 2.2374 | 0.08 | 500 | 2.3915 |
74
- | 2.2968 | 0.08 | 525 | 2.3906 |
75
- | 2.1904 | 0.09 | 550 | 2.3883 |
76
- | 2.3101 | 0.09 | 575 | 2.3894 |
77
- | 2.1126 | 0.1 | 600 | 2.3899 |
78
- | 2.2092 | 0.1 | 625 | 2.3934 |
79
- | 2.3005 | 0.1 | 650 | 2.3903 |
80
- | 2.2779 | 0.11 | 675 | 2.3876 |
81
- | 2.2523 | 0.11 | 700 | 2.3886 |
82
- | 2.2307 | 0.12 | 725 | 2.3879 |
83
- | 2.1317 | 0.12 | 750 | 2.3832 |
84
- | 2.1893 | 0.12 | 775 | 2.3848 |
85
- | 2.2732 | 0.13 | 800 | 2.3822 |
86
- | 2.2914 | 0.13 | 825 | 2.3855 |
87
- | 2.2633 | 0.14 | 850 | 2.3859 |
88
- | 2.1339 | 0.14 | 875 | 2.3866 |
89
- | 2.2065 | 0.14 | 900 | 2.3862 |
90
- | 2.1436 | 0.15 | 925 | 2.3807 |
91
- | 2.2782 | 0.15 | 950 | 2.3822 |
92
- | 2.2001 | 0.16 | 975 | 2.3800 |
93
- | 2.2608 | 0.16 | 1000 | 2.3785 |
94
- | 2.2654 | 0.16 | 1025 | 2.3817 |
95
- | 2.2662 | 0.17 | 1050 | 2.3783 |
96
- | 2.3061 | 0.17 | 1075 | 2.3777 |
97
- | 2.2297 | 0.18 | 1100 | 2.3763 |
98
- | 2.2705 | 0.18 | 1125 | 2.3775 |
99
- | 2.2261 | 0.18 | 1150 | 2.3759 |
100
- | 2.2812 | 0.19 | 1175 | 2.3773 |
101
- | 2.1363 | 0.19 | 1200 | 2.3753 |
102
- | 2.2382 | 0.2 | 1225 | 2.3756 |
103
- | 2.2064 | 0.2 | 1250 | 2.3744 |
104
- | 2.2559 | 0.2 | 1275 | 2.3698 |
105
- | 2.2875 | 0.21 | 1300 | 2.3730 |
106
- | 2.2541 | 0.21 | 1325 | 2.3737 |
107
- | 2.1415 | 0.22 | 1350 | 2.3732 |
108
- | 2.2529 | 0.22 | 1375 | 2.3721 |
109
- | 2.2271 | 0.22 | 1400 | 2.3752 |
110
- | 2.1849 | 0.23 | 1425 | 2.3738 |
111
- | 2.1707 | 0.23 | 1450 | 2.3728 |
112
- | 2.1363 | 0.24 | 1475 | 2.3729 |
113
- | 2.1778 | 0.24 | 1500 | 2.3731 |
114
- | 2.1146 | 0.24 | 1525 | 2.3795 |
115
- | 2.1843 | 0.25 | 1550 | 2.3775 |
116
- | 2.3094 | 0.25 | 1575 | 2.3727 |
117
- | 2.2488 | 0.26 | 1600 | 2.3758 |
118
- | 2.226 | 0.26 | 1625 | 2.3723 |
119
- | 2.3067 | 0.26 | 1650 | 2.3734 |
120
- | 2.2167 | 0.27 | 1675 | 2.3760 |
121
- | 2.2466 | 0.27 | 1700 | 2.3750 |
122
- | 2.2446 | 0.28 | 1725 | 2.3775 |
123
- | 2.2268 | 0.28 | 1750 | 2.3741 |
124
- | 2.2113 | 0.28 | 1775 | 2.3733 |
125
- | 2.1608 | 0.29 | 1800 | 2.3762 |
126
- | 2.2354 | 0.29 | 1825 | 2.3758 |
127
- | 2.2433 | 0.3 | 1850 | 2.3745 |
128
- | 2.2266 | 0.3 | 1875 | 2.3769 |
129
- | 2.2453 | 0.3 | 1900 | 2.3726 |
130
- | 2.3001 | 0.31 | 1925 | 2.3713 |
131
- | 2.2447 | 0.31 | 1950 | 2.3722 |
132
- | 2.2708 | 0.32 | 1975 | 2.3730 |
133
- | 2.1878 | 0.32 | 2000 | 2.3743 |
134
- | 2.2041 | 0.32 | 2025 | 2.3751 |
135
- | 2.1935 | 0.33 | 2050 | 2.3750 |
136
- | 2.1981 | 0.33 | 2075 | 2.3744 |
137
- | 2.2777 | 0.34 | 2100 | 2.3720 |
138
- | 2.3121 | 0.34 | 2125 | 2.3725 |
139
- | 2.2294 | 0.34 | 2150 | 2.3750 |
140
- | 2.1802 | 0.35 | 2175 | 2.3772 |
141
- | 2.214 | 0.35 | 2200 | 2.3738 |
142
- | 2.1631 | 0.36 | 2225 | 2.3740 |
143
- | 2.1546 | 0.36 | 2250 | 2.3764 |
144
- | 2.2841 | 0.36 | 2275 | 2.3743 |
145
- | 2.271 | 0.37 | 2300 | 2.3707 |
146
- | 2.1627 | 0.37 | 2325 | 2.3719 |
147
- | 2.2071 | 0.38 | 2350 | 2.3678 |
148
- | 2.2423 | 0.38 | 2375 | 2.3703 |
149
- | 2.2554 | 0.38 | 2400 | 2.3700 |
150
- | 2.1057 | 0.39 | 2425 | 2.3720 |
151
- | 2.0983 | 0.39 | 2450 | 2.3690 |
152
- | 2.1844 | 0.4 | 2475 | 2.3686 |
153
- | 2.2797 | 0.4 | 2500 | 2.3719 |
154
- | 2.2749 | 0.4 | 2525 | 2.3707 |
155
- | 2.1326 | 0.41 | 2550 | 2.3728 |
156
- | 2.1461 | 0.41 | 2575 | 2.3693 |
157
- | 2.2324 | 0.42 | 2600 | 2.3699 |
158
- | 2.2412 | 0.42 | 2625 | 2.3690 |
159
- | 2.28 | 0.42 | 2650 | 2.3696 |
160
- | 2.261 | 0.43 | 2675 | 2.3666 |
161
- | 2.2737 | 0.43 | 2700 | 2.3674 |
162
- | 2.2653 | 0.44 | 2725 | 2.3671 |
163
- | 2.2269 | 0.44 | 2750 | 2.3643 |
164
- | 2.245 | 0.44 | 2775 | 2.3641 |
165
- | 2.3077 | 0.45 | 2800 | 2.3665 |
166
- | 2.2143 | 0.45 | 2825 | 2.3667 |
167
- | 2.2595 | 0.46 | 2850 | 2.3662 |
168
- | 2.1638 | 0.46 | 2875 | 2.3661 |
169
- | 2.1935 | 0.46 | 2900 | 2.3645 |
170
- | 2.2063 | 0.47 | 2925 | 2.3659 |
171
- | 2.2755 | 0.47 | 2950 | 2.3664 |
172
- | 2.1977 | 0.48 | 2975 | 2.3649 |
173
- | 2.2519 | 0.48 | 3000 | 2.3616 |
174
- | 2.3353 | 0.48 | 3025 | 2.3648 |
175
- | 2.223 | 0.49 | 3050 | 2.3629 |
176
- | 2.2614 | 0.49 | 3075 | 2.3611 |
177
- | 2.2983 | 0.5 | 3100 | 2.3663 |
178
- | 2.1907 | 0.5 | 3125 | 2.3650 |
179
- | 2.2683 | 0.5 | 3150 | 2.3629 |
180
- | 2.1609 | 0.51 | 3175 | 2.3633 |
181
- | 2.2316 | 0.51 | 3200 | 2.3626 |
182
- | 2.1589 | 0.52 | 3225 | 2.3603 |
183
- | 2.1479 | 0.52 | 3250 | 2.3598 |
184
- | 2.2401 | 0.52 | 3275 | 2.3617 |
185
- | 2.2073 | 0.53 | 3300 | 2.3608 |
186
- | 2.094 | 0.53 | 3325 | 2.3609 |
187
- | 2.2297 | 0.54 | 3350 | 2.3592 |
188
- | 2.1305 | 0.54 | 3375 | 2.3585 |
189
- | 2.1517 | 0.54 | 3400 | 2.3598 |
190
- | 2.1592 | 0.55 | 3425 | 2.3626 |
191
- | 2.0812 | 0.55 | 3450 | 2.3636 |
192
- | 2.219 | 0.56 | 3475 | 2.3633 |
193
- | 2.2632 | 0.56 | 3500 | 2.3625 |
194
- | 2.2302 | 0.56 | 3525 | 2.3616 |
195
- | 2.1926 | 0.57 | 3550 | 2.3623 |
196
- | 2.1878 | 0.57 | 3575 | 2.3630 |
197
- | 2.3519 | 0.58 | 3600 | 2.3609 |
198
- | 2.1699 | 0.58 | 3625 | 2.3599 |
199
- | 2.3576 | 0.58 | 3650 | 2.3618 |
200
- | 2.1629 | 0.59 | 3675 | 2.3637 |
201
- | 2.2982 | 0.59 | 3700 | 2.3606 |
202
- | 2.1949 | 0.6 | 3725 | 2.3649 |
203
- | 2.135 | 0.6 | 3750 | 2.3618 |
204
- | 2.0752 | 0.6 | 3775 | 2.3637 |
205
- | 2.2786 | 0.61 | 3800 | 2.3643 |
206
- | 2.0974 | 0.61 | 3825 | 2.3644 |
207
- | 2.2097 | 0.62 | 3850 | 2.3621 |
208
- | 2.1372 | 0.62 | 3875 | 2.3642 |
209
- | 2.2502 | 0.62 | 3900 | 2.3652 |
210
- | 2.1817 | 0.63 | 3925 | 2.3613 |
211
- | 2.1891 | 0.63 | 3950 | 2.3659 |
212
- | 2.2261 | 0.64 | 3975 | 2.3630 |
213
- | 2.2826 | 0.64 | 4000 | 2.3591 |
214
- | 2.2308 | 0.64 | 4025 | 2.3617 |
215
- | 2.1944 | 0.65 | 4050 | 2.3615 |
216
- | 2.1638 | 0.65 | 4075 | 2.3644 |
217
- | 2.1741 | 0.66 | 4100 | 2.3600 |
218
- | 2.2092 | 0.66 | 4125 | 2.3602 |
219
- | 2.1921 | 0.66 | 4150 | 2.3610 |
220
- | 2.1029 | 0.67 | 4175 | 2.3604 |
221
- | 2.2553 | 0.67 | 4200 | 2.3578 |
222
- | 2.1924 | 0.68 | 4225 | 2.3625 |
223
- | 2.1914 | 0.68 | 4250 | 2.3619 |
224
- | 2.2556 | 0.68 | 4275 | 2.3583 |
225
- | 2.188 | 0.69 | 4300 | 2.3600 |
226
- | 2.2339 | 0.69 | 4325 | 2.3594 |
227
- | 2.2484 | 0.7 | 4350 | 2.3593 |
228
- | 2.2761 | 0.7 | 4375 | 2.3585 |
229
- | 2.2305 | 0.7 | 4400 | 2.3606 |
230
- | 2.2434 | 0.71 | 4425 | 2.3586 |
231
- | 2.2879 | 0.71 | 4450 | 2.3586 |
232
- | 2.1643 | 0.72 | 4475 | 2.3639 |
233
- | 2.1697 | 0.72 | 4500 | 2.3573 |
234
- | 2.2665 | 0.72 | 4525 | 2.3574 |
235
- | 2.2349 | 0.73 | 4550 | 2.3593 |
236
- | 2.1459 | 0.73 | 4575 | 2.3554 |
237
- | 2.1619 | 0.74 | 4600 | 2.3546 |
238
- | 2.1292 | 0.74 | 4625 | 2.3542 |
239
- | 2.3202 | 0.74 | 4650 | 2.3591 |
240
- | 2.2165 | 0.75 | 4675 | 2.3562 |
241
- | 2.233 | 0.75 | 4700 | 2.3566 |
242
- | 2.2378 | 0.76 | 4725 | 2.3574 |
243
- | 2.1236 | 0.76 | 4750 | 2.3554 |
244
- | 2.2054 | 0.76 | 4775 | 2.3586 |
245
- | 2.1807 | 0.77 | 4800 | 2.3543 |
246
- | 2.2965 | 0.77 | 4825 | 2.3570 |
247
- | 2.3543 | 0.78 | 4850 | 2.3562 |
248
- | 2.1981 | 0.78 | 4875 | 2.3575 |
249
- | 2.1156 | 0.78 | 4900 | 2.3586 |
250
- | 2.1352 | 0.79 | 4925 | 2.3560 |
251
- | 2.3162 | 0.79 | 4950 | 2.3607 |
252
- | 2.1469 | 0.8 | 4975 | 2.3575 |
253
- | 2.2243 | 0.8 | 5000 | 2.3591 |
254
 
255
 
256
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the None dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 2.3402
19
 
20
  ## Model description
21
 
 
45
  - total_eval_batch_size: 4
46
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
47
  - lr_scheduler_type: linear
48
+ - training_steps: 2500
49
 
50
  ### Training results
51
 
52
  | Training Loss | Epoch | Step | Validation Loss |
53
  |:-------------:|:-----:|:----:|:---------------:|
54
+ | 3.7252 | 0.01 | 50 | 2.3893 |
55
+ | 2.2531 | 0.02 | 100 | 2.4723 |
56
+ | 2.32 | 0.02 | 150 | 2.4385 |
57
+ | 2.2363 | 0.03 | 200 | 2.4210 |
58
+ | 2.3078 | 0.04 | 250 | 2.4118 |
59
+ | 2.2389 | 0.05 | 300 | 2.4025 |
60
+ | 2.0902 | 0.06 | 350 | 2.3984 |
61
+ | 2.2878 | 0.06 | 400 | 2.3965 |
62
+ | 2.2485 | 0.07 | 450 | 2.3924 |
63
+ | 2.2375 | 0.08 | 500 | 2.3895 |
64
+ | 2.1901 | 0.09 | 550 | 2.3909 |
65
+ | 2.1128 | 0.1 | 600 | 2.3886 |
66
+ | 2.2983 | 0.1 | 650 | 2.3892 |
67
+ | 2.2547 | 0.11 | 700 | 2.3873 |
68
+ | 2.1322 | 0.12 | 750 | 2.3861 |
69
+ | 2.2715 | 0.13 | 800 | 2.3827 |
70
+ | 2.263 | 0.14 | 850 | 2.3845 |
71
+ | 2.2066 | 0.14 | 900 | 2.3836 |
72
+ | 2.2781 | 0.15 | 950 | 2.3837 |
73
+ | 2.2597 | 0.16 | 1000 | 2.3778 |
74
+ | 2.2642 | 0.17 | 1050 | 2.3764 |
75
+ | 2.2296 | 0.18 | 1100 | 2.3805 |
76
+ | 2.2289 | 0.18 | 1150 | 2.3784 |
77
+ | 2.1372 | 0.19 | 1200 | 2.3773 |
78
+ | 2.2059 | 0.2 | 1250 | 2.3732 |
79
+ | 2.2847 | 0.21 | 1300 | 2.3719 |
80
+ | 2.1404 | 0.22 | 1350 | 2.3739 |
81
+ | 2.2261 | 0.22 | 1400 | 2.3752 |
82
+ | 2.1713 | 0.23 | 1450 | 2.3750 |
83
+ | 2.1787 | 0.24 | 1500 | 2.3732 |
84
+ | 2.1866 | 0.25 | 1550 | 2.3759 |
85
+ | 2.2471 | 0.26 | 1600 | 2.3760 |
86
+ | 2.307 | 0.26 | 1650 | 2.3745 |
87
+ | 2.2457 | 0.27 | 1700 | 2.3746 |
88
+ | 2.2265 | 0.28 | 1750 | 2.3775 |
89
+ | 2.163 | 0.29 | 1800 | 2.3797 |
90
+ | 2.2411 | 0.3 | 1850 | 2.3760 |
91
+ | 2.247 | 0.3 | 1900 | 2.3770 |
92
+ | 2.2449 | 0.31 | 1950 | 2.3749 |
93
+ | 2.1884 | 0.32 | 2000 | 2.3728 |
94
+ | 2.1909 | 0.33 | 2050 | 2.3770 |
95
+ | 2.2813 | 0.34 | 2100 | 2.3773 |
96
+ | 2.2306 | 0.34 | 2150 | 2.3755 |
97
+ | 2.2158 | 0.35 | 2200 | 2.3777 |
98
+ | 2.1557 | 0.36 | 2250 | 2.3783 |
99
+ | 2.2715 | 0.37 | 2300 | 2.3704 |
100
+ | 2.2053 | 0.38 | 2350 | 2.3729 |
101
+ | 2.2541 | 0.38 | 2400 | 2.3715 |
102
+ | 2.0971 | 0.39 | 2450 | 2.3747 |
103
+ | 2.2791 | 0.4 | 2500 | 2.3727 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
104
 
105
 
106
  ### Framework versions
model-00001-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:69b63967bf6e9412919957341ef1a5cc8d78e4cb0c815670d36272e72ddc15cd
3
  size 4943163992
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7c216e3f0feb4aed8bc1135edf5f1999563f11c01ece4ef3c17a7018d593d419
3
  size 4943163992
model-00002-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d9a7de3f5243a7e4909935bb959c8af826421aba6e2051934d2a601b9f818234
3
  size 4999821144
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:05c6e97698de854abbc47309e5db0ecaa49da886c248208b2bbdb245b46d23d7
3
  size 4999821144
model-00003-of-00003.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:7cd298af76ce768f3e57280a90832b59376d18111c31b5d9db2cf87a1f828963
3
  size 4540517840
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63e02e1b7360b79190b22246938831e6ee965cd8096a7cc191f84e2ede2c4455
3
  size 4540517840