Hugging Face's logo

RichardErkhov
/
thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf

GGUF
Model card Files Community
RichardErkhov commited on
Commit
2573f94
•
1 Parent(s): 59eedd4

uploaded readme

Browse files
Files changed (1) hide show
  1. README.md +199 -0
README.md ADDED
@@ -0,0 +1,199 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ Quantization made by Richard Erkhov.
2
+
3
+ [Github](https://github.com/RichardErkhov)
4
+
5
+ [Discord](https://discord.gg/pvy7H8DZMG)
6
+
7
+ [Request more models](https://github.com/RichardErkhov/quant_request)
8
+
9
+
10
+ sparse_mistral_7b_refined_web_50p_2024-04-13 - GGUF
11
+ - Model creator: https://huggingface.co/thrunlab/
12
+ - Original model: https://huggingface.co/thrunlab/sparse_mistral_7b_refined_web_50p_2024-04-13/
13
+
14
+
15
+ | Name | Quant method | Size |
16
+ | ---- | ---- | ---- |
17
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q2_K.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q2_K.gguf) | Q2_K | 2.53GB |
18
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.IQ3_XS.gguf) | IQ3_XS | 2.81GB |
19
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.IQ3_S.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.IQ3_S.gguf) | IQ3_S | 2.96GB |
20
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K_S.gguf) | Q3_K_S | 2.95GB |
21
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.IQ3_M.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.IQ3_M.gguf) | IQ3_M | 3.06GB |
22
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K.gguf) | Q3_K | 3.28GB |
23
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K_M.gguf) | Q3_K_M | 3.28GB |
24
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q3_K_L.gguf) | Q3_K_L | 3.56GB |
25
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.IQ4_XS.gguf) | IQ4_XS | 3.67GB |
26
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_0.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_0.gguf) | Q4_0 | 3.83GB |
27
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.IQ4_NL.gguf) | IQ4_NL | 3.87GB |
28
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_K_S.gguf) | Q4_K_S | 3.86GB |
29
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_K.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_K.gguf) | Q4_K | 4.07GB |
30
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_K_M.gguf) | Q4_K_M | 4.07GB |
31
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_1.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q4_1.gguf) | Q4_1 | 4.24GB |
32
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_0.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_0.gguf) | Q5_0 | 4.65GB |
33
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_K_S.gguf) | Q5_K_S | 4.65GB |
34
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_K.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_K.gguf) | Q5_K | 4.78GB |
35
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_K_M.gguf) | Q5_K_M | 4.78GB |
36
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_1.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q5_1.gguf) | Q5_1 | 5.07GB |
37
+ | [sparse_mistral_7b_refined_web_50p_2024-04-13.Q6_K.gguf](https://huggingface.co/RichardErkhov/thrunlab_-_sparse_mistral_7b_refined_web_50p_2024-04-13-gguf/blob/main/sparse_mistral_7b_refined_web_50p_2024-04-13.Q6_K.gguf) | Q6_K | 5.53GB |
38
+
39
+
40
+
41
+
42
+ Original model description:
43
+ ---
44
+ license: apache-2.0
45
+ base_model: mistralai/Mistral-7B-v0.1
46
+ tags:
47
+ - generated_from_trainer
48
+ model-index:
49
+ - name: sparse_mistral_7b_refined_web_50p_2024-04-13
50
+ results: []
51
+ ---
52
+
53
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
54
+ should probably proofread and complete it, then remove this comment. -->
55
+
56
+ # sparse_mistral_7b_refined_web_50p_2024-04-13
57
+
58
+ 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.
59
+ It achieves the following results on the evaluation set:
60
+ - Loss: 2.1985
61
+
62
+ ## Model description
63
+
64
+ More information needed
65
+
66
+ ## Intended uses & limitations
67
+
68
+ More information needed
69
+
70
+ ## Training and evaluation data
71
+
72
+ More information needed
73
+
74
+ ## Training procedure
75
+
76
+ ### Training hyperparameters
77
+
78
+ The following hyperparameters were used during training:
79
+ - learning_rate: 1e-05
80
+ - train_batch_size: 1
81
+ - eval_batch_size: 4
82
+ - seed: 0
83
+ - distributed_type: multi-GPU
84
+ - num_devices: 4
85
+ - gradient_accumulation_steps: 8
86
+ - total_train_batch_size: 32
87
+ - total_eval_batch_size: 16
88
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
89
+ - lr_scheduler_type: linear
90
+ - training_steps: 2350
91
+
92
+ ### Training results
93
+
94
+ | Training Loss | Epoch | Step | Validation Loss |
95
+ |:-------------:|:-----:|:----:|:---------------:|
96
+ | 2.3391 | 0.01 | 25 | 2.4196 |
97
+ | 2.2711 | 0.02 | 50 | 2.3577 |
98
+ | 2.3054 | 0.02 | 75 | 2.3158 |
99
+ | 2.2795 | 0.03 | 100 | 2.2966 |
100
+ | 2.3175 | 0.04 | 125 | 2.2846 |
101
+ | 2.2388 | 0.05 | 150 | 2.2766 |
102
+ | 2.1679 | 0.06 | 175 | 2.2705 |
103
+ | 2.2996 | 0.06 | 200 | 2.2678 |
104
+ | 2.2788 | 0.07 | 225 | 2.2647 |
105
+ | 2.2448 | 0.08 | 250 | 2.2637 |
106
+ | 2.1837 | 0.09 | 275 | 2.2624 |
107
+ | 2.2089 | 0.1 | 300 | 2.2621 |
108
+ | 2.2686 | 0.1 | 325 | 2.2601 |
109
+ | 2.2254 | 0.11 | 350 | 2.2593 |
110
+ | 2.162 | 0.12 | 375 | 2.2590 |
111
+ | 2.2687 | 0.13 | 400 | 2.2563 |
112
+ | 2.2595 | 0.14 | 425 | 2.2571 |
113
+ | 2.186 | 0.14 | 450 | 2.2564 |
114
+ | 2.2689 | 0.15 | 475 | 2.2580 |
115
+ | 2.2472 | 0.16 | 500 | 2.2554 |
116
+ | 2.2005 | 0.17 | 525 | 2.2553 |
117
+ | 2.1983 | 0.18 | 550 | 2.2552 |
118
+ | 2.2388 | 0.18 | 575 | 2.2547 |
119
+ | 2.1443 | 0.19 | 600 | 2.2555 |
120
+ | 2.2198 | 0.2 | 625 | 2.2534 |
121
+ | 2.3008 | 0.21 | 650 | 2.2536 |
122
+ | 2.179 | 0.22 | 675 | 2.2521 |
123
+ | 2.2069 | 0.22 | 700 | 2.2531 |
124
+ | 2.1819 | 0.23 | 725 | 2.2526 |
125
+ | 2.1218 | 0.24 | 750 | 2.2536 |
126
+ | 2.1845 | 0.25 | 775 | 2.2515 |
127
+ | 2.2167 | 0.26 | 800 | 2.2510 |
128
+ | 2.2252 | 0.26 | 825 | 2.2520 |
129
+ | 2.1664 | 0.27 | 850 | 2.2519 |
130
+ | 2.1853 | 0.28 | 875 | 2.2530 |
131
+ | 2.1499 | 0.29 | 900 | 2.2513 |
132
+ | 2.2763 | 0.3 | 925 | 2.2517 |
133
+ | 2.2528 | 0.3 | 950 | 2.2518 |
134
+ | 2.2505 | 0.31 | 975 | 2.2500 |
135
+ | 2.1683 | 0.32 | 1000 | 2.2502 |
136
+ | 2.2177 | 0.33 | 1025 | 2.2501 |
137
+ | 2.238 | 0.34 | 1050 | 2.2516 |
138
+ | 2.193 | 0.34 | 1075 | 2.2507 |
139
+ | 2.2025 | 0.35 | 1100 | 2.2502 |
140
+ | 2.0944 | 0.36 | 1125 | 2.2512 |
141
+ | 2.2272 | 0.37 | 1150 | 2.2508 |
142
+ | 2.2264 | 0.38 | 1175 | 2.2500 |
143
+ | 2.1837 | 0.38 | 1200 | 2.2507 |
144
+ | 2.1444 | 0.39 | 1225 | 2.2489 |
145
+ | 2.2464 | 0.4 | 1250 | 2.2499 |
146
+ | 2.1388 | 0.41 | 1275 | 2.2508 |
147
+ | 2.193 | 0.42 | 1300 | 2.2492 |
148
+ | 2.2376 | 0.42 | 1325 | 2.2506 |
149
+ | 2.2212 | 0.43 | 1350 | 2.2478 |
150
+ | 2.2002 | 0.44 | 1375 | 2.2488 |
151
+ | 2.2729 | 0.45 | 1400 | 2.2484 |
152
+ | 2.2329 | 0.46 | 1425 | 2.2473 |
153
+ | 2.1919 | 0.46 | 1450 | 2.2481 |
154
+ | 2.2102 | 0.47 | 1475 | 2.2475 |
155
+ | 2.1466 | 0.48 | 1500 | 2.2473 |
156
+ | 2.1819 | 0.49 | 1525 | 2.2478 |
157
+ | 2.2558 | 0.5 | 1550 | 2.2468 |
158
+ | 2.2137 | 0.5 | 1575 | 2.2463 |
159
+ | 2.2288 | 0.51 | 1600 | 2.2466 |
160
+ | 2.1479 | 0.52 | 1625 | 2.2468 |
161
+ | 2.1726 | 0.53 | 1650 | 2.2471 |
162
+ | 2.1805 | 0.54 | 1675 | 2.2454 |
163
+ | 2.1505 | 0.54 | 1700 | 2.2470 |
164
+ | 2.1337 | 0.55 | 1725 | 2.2465 |
165
+ | 2.2413 | 0.56 | 1750 | 2.2460 |
166
+ | 2.152 | 0.57 | 1775 | 2.2478 |
167
+ | 2.2669 | 0.58 | 1800 | 2.2471 |
168
+ | 2.2925 | 0.58 | 1825 | 2.2465 |
169
+ | 2.222 | 0.59 | 1850 | 2.2457 |
170
+ | 2.1308 | 0.6 | 1875 | 2.2466 |
171
+ | 2.201 | 0.61 | 1900 | 2.2456 |
172
+ | 2.2247 | 0.62 | 1925 | 2.2460 |
173
+ | 2.2426 | 0.62 | 1950 | 2.2463 |
174
+ | 2.2312 | 0.63 | 1975 | 2.2465 |
175
+ | 2.2679 | 0.64 | 2000 | 2.2464 |
176
+ | 2.1928 | 0.65 | 2025 | 2.2463 |
177
+ | 2.2087 | 0.66 | 2050 | 2.2455 |
178
+ | 2.1792 | 0.66 | 2075 | 2.2470 |
179
+ | 2.252 | 0.67 | 2100 | 2.2468 |
180
+ | 2.2018 | 0.68 | 2125 | 2.2456 |
181
+ | 2.2006 | 0.69 | 2150 | 2.2451 |
182
+ | 2.2076 | 0.7 | 2175 | 2.2449 |
183
+ | 2.2436 | 0.7 | 2200 | 2.2460 |
184
+ | 2.2156 | 0.71 | 2225 | 2.2477 |
185
+ | 2.1348 | 0.72 | 2250 | 2.2455 |
186
+ | 2.1338 | 0.73 | 2275 | 2.2450 |
187
+ | 2.2147 | 0.74 | 2300 | 2.2455 |
188
+ | 2.2766 | 0.74 | 2325 | 2.2444 |
189
+ | 2.204 | 0.75 | 2350 | 2.2458 |
190
+
191
+
192
+ ### Framework versions
193
+
194
+ - Transformers 4.36.2
195
+ - Pytorch 2.1.2+cu121
196
+ - Datasets 2.15.0
197
+ - Tokenizers 0.15.0
198
+
199
+