End of training
Browse files
README.md
CHANGED
|
@@ -18,7 +18,7 @@ should probably proofread and complete it, then remove this comment. -->
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on the None dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
-
- Loss: 0.
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
@@ -50,170 +50,184 @@ The following hyperparameters were used during training:
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
|:-------------:|:------:|:-----:|:---------------:|
|
| 53 |
-
| 0.
|
| 54 |
-
| 0.
|
| 55 |
-
| 0.
|
| 56 |
-
| 0.
|
| 57 |
-
| 0.
|
| 58 |
-
| 0.
|
| 59 |
-
| 0.
|
| 60 |
-
| 0.
|
| 61 |
-
| 0.
|
| 62 |
-
| 0.
|
| 63 |
-
| 0.
|
| 64 |
-
| 0.
|
| 65 |
-
| 0.
|
| 66 |
-
| 0.
|
| 67 |
-
| 0.
|
| 68 |
-
| 0.
|
| 69 |
-
| 0.
|
| 70 |
-
| 0.
|
| 71 |
-
| 0.
|
| 72 |
-
| 0.
|
| 73 |
-
| 0.
|
| 74 |
-
| 0.
|
| 75 |
-
| 0.
|
| 76 |
-
| 0.
|
| 77 |
-
| 0.
|
| 78 |
-
| 0.
|
| 79 |
-
| 0.
|
| 80 |
-
| 0.
|
| 81 |
-
| 0.
|
| 82 |
-
| 0.
|
| 83 |
-
| 0.
|
| 84 |
-
| 0.
|
| 85 |
-
| 0.
|
| 86 |
-
| 0.
|
| 87 |
-
| 0.
|
| 88 |
-
| 0.
|
| 89 |
-
| 0.
|
| 90 |
-
| 0.
|
| 91 |
-
| 0.
|
| 92 |
-
| 0.
|
| 93 |
-
| 0.
|
| 94 |
-
| 0.
|
| 95 |
-
| 0.
|
| 96 |
-
| 0.
|
| 97 |
-
| 0.
|
| 98 |
-
| 0.
|
| 99 |
-
| 0.
|
| 100 |
-
| 0.
|
| 101 |
-
| 0.
|
| 102 |
-
| 0.
|
| 103 |
-
| 0.
|
| 104 |
-
| 0.
|
| 105 |
-
| 0.
|
| 106 |
-
| 0.
|
| 107 |
-
| 0.
|
| 108 |
-
| 0.
|
| 109 |
-
| 0.
|
| 110 |
-
| 0.
|
| 111 |
-
| 0.
|
| 112 |
-
| 0.
|
| 113 |
-
| 0.
|
| 114 |
-
| 0.
|
| 115 |
-
| 0.
|
| 116 |
-
| 0.
|
| 117 |
-
| 0.
|
| 118 |
-
| 0.
|
| 119 |
-
| 0.
|
| 120 |
-
| 0.
|
| 121 |
-
| 0.
|
| 122 |
-
| 0.
|
| 123 |
-
| 0.
|
| 124 |
-
| 0.
|
| 125 |
-
| 0.
|
| 126 |
-
| 0.
|
| 127 |
-
| 0.
|
| 128 |
-
| 0.
|
| 129 |
-
| 0.
|
| 130 |
-
| 0.
|
| 131 |
-
| 0.
|
| 132 |
-
| 0.
|
| 133 |
-
| 0.
|
| 134 |
-
| 0.
|
| 135 |
-
| 0.
|
| 136 |
-
| 0.
|
| 137 |
-
| 0.
|
| 138 |
-
| 0.
|
| 139 |
-
| 0.
|
| 140 |
-
| 0.
|
| 141 |
-
| 0.
|
| 142 |
-
| 0.
|
| 143 |
-
| 0.
|
| 144 |
-
| 0.
|
| 145 |
-
| 0.
|
| 146 |
-
| 0.
|
| 147 |
-
| 0.
|
| 148 |
-
| 0.
|
| 149 |
-
| 0.
|
| 150 |
-
| 0.
|
| 151 |
-
| 0.
|
| 152 |
-
| 0.
|
| 153 |
-
| 0.
|
| 154 |
-
| 0.
|
| 155 |
-
| 0.
|
| 156 |
-
| 0.
|
| 157 |
-
| 0.
|
| 158 |
-
| 0.
|
| 159 |
-
| 0.
|
| 160 |
-
| 0.
|
| 161 |
-
| 0.
|
| 162 |
-
| 0.
|
| 163 |
-
| 0.
|
| 164 |
-
| 0.
|
| 165 |
-
| 0.
|
| 166 |
-
| 0.
|
| 167 |
-
| 0.
|
| 168 |
-
| 0.
|
| 169 |
-
| 0.
|
| 170 |
-
| 0.
|
| 171 |
-
| 0.
|
| 172 |
-
| 0.
|
| 173 |
-
| 0.
|
| 174 |
-
| 0.
|
| 175 |
-
| 0.
|
| 176 |
-
| 0.
|
| 177 |
-
| 0.
|
| 178 |
-
| 0.
|
| 179 |
-
| 0.
|
| 180 |
-
| 0.
|
| 181 |
-
| 0.
|
| 182 |
-
| 0.
|
| 183 |
-
| 0.
|
| 184 |
-
| 0.
|
| 185 |
-
| 0.
|
| 186 |
-
| 0.
|
| 187 |
-
| 0.
|
| 188 |
-
| 0.
|
| 189 |
-
| 0.
|
| 190 |
-
| 0.
|
| 191 |
-
| 0.
|
| 192 |
-
| 0.
|
| 193 |
-
| 0.
|
| 194 |
-
| 0.
|
| 195 |
-
| 0.
|
| 196 |
-
| 0.
|
| 197 |
-
| 0.
|
| 198 |
-
| 0.
|
| 199 |
-
| 0.
|
| 200 |
-
| 0.
|
| 201 |
-
| 0.
|
| 202 |
-
| 0.
|
| 203 |
-
| 0.
|
| 204 |
-
| 0.
|
| 205 |
-
| 0.
|
| 206 |
-
| 0.
|
| 207 |
-
| 0.
|
| 208 |
-
| 0.
|
| 209 |
-
| 0.
|
| 210 |
-
| 0.
|
| 211 |
-
| 0.
|
| 212 |
-
| 0.
|
| 213 |
-
| 0.
|
| 214 |
-
| 0.
|
| 215 |
-
| 0.
|
| 216 |
-
| 0.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 217 |
|
| 218 |
|
| 219 |
### Framework versions
|
|
|
|
| 18 |
|
| 19 |
This model is a fine-tuned version of [codellama/CodeLlama-7b-Instruct-hf](https://huggingface.co/codellama/CodeLlama-7b-Instruct-hf) on the None dataset.
|
| 20 |
It achieves the following results on the evaluation set:
|
| 21 |
+
- Loss: 0.3001
|
| 22 |
|
| 23 |
## Model description
|
| 24 |
|
|
|
|
| 50 |
|
| 51 |
| Training Loss | Epoch | Step | Validation Loss |
|
| 52 |
|:-------------:|:------:|:-----:|:---------------:|
|
| 53 |
+
| 0.1811 | 2.3788 | 13000 | 0.2900 |
|
| 54 |
+
| 0.2081 | 2.3879 | 13050 | 0.2917 |
|
| 55 |
+
| 0.1862 | 2.3971 | 13100 | 0.2925 |
|
| 56 |
+
| 0.1806 | 2.4062 | 13150 | 0.2905 |
|
| 57 |
+
| 0.1844 | 2.4154 | 13200 | 0.2887 |
|
| 58 |
+
| 0.1862 | 2.4245 | 13250 | 0.2879 |
|
| 59 |
+
| 0.1783 | 2.4337 | 13300 | 0.2897 |
|
| 60 |
+
| 0.2143 | 2.4428 | 13350 | 0.2871 |
|
| 61 |
+
| 0.2068 | 2.4520 | 13400 | 0.2877 |
|
| 62 |
+
| 0.1815 | 2.4611 | 13450 | 0.2877 |
|
| 63 |
+
| 0.1657 | 2.4703 | 13500 | 0.2871 |
|
| 64 |
+
| 0.2198 | 2.4794 | 13550 | 0.2857 |
|
| 65 |
+
| 0.1709 | 2.4886 | 13600 | 0.2872 |
|
| 66 |
+
| 0.1741 | 2.4977 | 13650 | 0.2857 |
|
| 67 |
+
| 0.217 | 2.5069 | 13700 | 0.2847 |
|
| 68 |
+
| 0.2064 | 2.5160 | 13750 | 0.2861 |
|
| 69 |
+
| 0.1813 | 2.5252 | 13800 | 0.2856 |
|
| 70 |
+
| 0.1895 | 2.5343 | 13850 | 0.2867 |
|
| 71 |
+
| 0.1881 | 2.5435 | 13900 | 0.2856 |
|
| 72 |
+
| 0.1815 | 2.5526 | 13950 | 0.2872 |
|
| 73 |
+
| 0.1772 | 2.5618 | 14000 | 0.2849 |
|
| 74 |
+
| 0.1808 | 2.5709 | 14050 | 0.2858 |
|
| 75 |
+
| 0.1808 | 2.5801 | 14100 | 0.2851 |
|
| 76 |
+
| 0.1949 | 2.5892 | 14150 | 0.2865 |
|
| 77 |
+
| 0.1894 | 2.5984 | 14200 | 0.2859 |
|
| 78 |
+
| 0.18 | 2.6075 | 14250 | 0.2846 |
|
| 79 |
+
| 0.1765 | 2.6167 | 14300 | 0.2842 |
|
| 80 |
+
| 0.178 | 2.6258 | 14350 | 0.2841 |
|
| 81 |
+
| 0.2042 | 2.6349 | 14400 | 0.2844 |
|
| 82 |
+
| 0.2073 | 2.6441 | 14450 | 0.2854 |
|
| 83 |
+
| 0.1748 | 2.6532 | 14500 | 0.2842 |
|
| 84 |
+
| 0.2015 | 2.6624 | 14550 | 0.2838 |
|
| 85 |
+
| 0.1968 | 2.6715 | 14600 | 0.2836 |
|
| 86 |
+
| 0.1919 | 2.6807 | 14650 | 0.2846 |
|
| 87 |
+
| 0.1811 | 2.6898 | 14700 | 0.2834 |
|
| 88 |
+
| 0.1751 | 2.6990 | 14750 | 0.2852 |
|
| 89 |
+
| 0.1889 | 2.7081 | 14800 | 0.2818 |
|
| 90 |
+
| 0.184 | 2.7173 | 14850 | 0.2833 |
|
| 91 |
+
| 0.1861 | 2.7264 | 14900 | 0.2852 |
|
| 92 |
+
| 0.1997 | 2.7356 | 14950 | 0.2835 |
|
| 93 |
+
| 0.1638 | 2.7447 | 15000 | 0.2839 |
|
| 94 |
+
| 0.1941 | 2.7539 | 15050 | 0.2830 |
|
| 95 |
+
| 0.1759 | 2.7630 | 15100 | 0.2833 |
|
| 96 |
+
| 0.1803 | 2.7722 | 15150 | 0.2837 |
|
| 97 |
+
| 0.1727 | 2.7813 | 15200 | 0.2834 |
|
| 98 |
+
| 0.2242 | 2.7905 | 15250 | 0.2816 |
|
| 99 |
+
| 0.1911 | 2.7996 | 15300 | 0.2849 |
|
| 100 |
+
| 0.196 | 2.8088 | 15350 | 0.2825 |
|
| 101 |
+
| 0.196 | 2.8179 | 15400 | 0.2815 |
|
| 102 |
+
| 0.1701 | 2.8271 | 15450 | 0.2800 |
|
| 103 |
+
| 0.1771 | 2.8362 | 15500 | 0.2823 |
|
| 104 |
+
| 0.187 | 2.8454 | 15550 | 0.2816 |
|
| 105 |
+
| 0.1804 | 2.8545 | 15600 | 0.2823 |
|
| 106 |
+
| 0.1722 | 2.8637 | 15650 | 0.2805 |
|
| 107 |
+
| 0.1957 | 2.8728 | 15700 | 0.2777 |
|
| 108 |
+
| 0.1725 | 2.8820 | 15750 | 0.2794 |
|
| 109 |
+
| 0.1945 | 2.8911 | 15800 | 0.2793 |
|
| 110 |
+
| 0.1884 | 2.9003 | 15850 | 0.2781 |
|
| 111 |
+
| 0.1687 | 2.9094 | 15900 | 0.2797 |
|
| 112 |
+
| 0.1965 | 2.9186 | 15950 | 0.2783 |
|
| 113 |
+
| 0.1669 | 2.9277 | 16000 | 0.2789 |
|
| 114 |
+
| 0.1827 | 2.9369 | 16050 | 0.2781 |
|
| 115 |
+
| 0.1663 | 2.9460 | 16100 | 0.2779 |
|
| 116 |
+
| 0.1665 | 2.9552 | 16150 | 0.2807 |
|
| 117 |
+
| 0.1812 | 2.9643 | 16200 | 0.2796 |
|
| 118 |
+
| 0.1995 | 2.9735 | 16250 | 0.2784 |
|
| 119 |
+
| 0.1792 | 2.9826 | 16300 | 0.2786 |
|
| 120 |
+
| 0.184 | 2.9918 | 16350 | 0.2786 |
|
| 121 |
+
| 0.1727 | 3.0009 | 16400 | 0.2781 |
|
| 122 |
+
| 0.1239 | 3.0101 | 16450 | 0.3034 |
|
| 123 |
+
| 0.1256 | 3.0192 | 16500 | 0.2975 |
|
| 124 |
+
| 0.1732 | 3.0284 | 16550 | 0.2993 |
|
| 125 |
+
| 0.1216 | 3.0375 | 16600 | 0.2999 |
|
| 126 |
+
| 0.1283 | 3.0467 | 16650 | 0.3015 |
|
| 127 |
+
| 0.1225 | 3.0558 | 16700 | 0.3011 |
|
| 128 |
+
| 0.1284 | 3.0650 | 16750 | 0.3038 |
|
| 129 |
+
| 0.1246 | 3.0741 | 16800 | 0.2990 |
|
| 130 |
+
| 0.1277 | 3.0833 | 16850 | 0.3011 |
|
| 131 |
+
| 0.1237 | 3.0924 | 16900 | 0.3023 |
|
| 132 |
+
| 0.1174 | 3.1016 | 16950 | 0.3053 |
|
| 133 |
+
| 0.1202 | 3.1107 | 17000 | 0.3037 |
|
| 134 |
+
| 0.121 | 3.1199 | 17050 | 0.3013 |
|
| 135 |
+
| 0.126 | 3.1290 | 17100 | 0.3034 |
|
| 136 |
+
| 0.1188 | 3.1382 | 17150 | 0.3008 |
|
| 137 |
+
| 0.1346 | 3.1473 | 17200 | 0.2997 |
|
| 138 |
+
| 0.1428 | 3.1565 | 17250 | 0.3011 |
|
| 139 |
+
| 0.1263 | 3.1656 | 17300 | 0.3038 |
|
| 140 |
+
| 0.1391 | 3.1747 | 17350 | 0.3044 |
|
| 141 |
+
| 0.128 | 3.1839 | 17400 | 0.3026 |
|
| 142 |
+
| 0.1367 | 3.1930 | 17450 | 0.3040 |
|
| 143 |
+
| 0.1214 | 3.2022 | 17500 | 0.3046 |
|
| 144 |
+
| 0.1253 | 3.2113 | 17550 | 0.3049 |
|
| 145 |
+
| 0.1157 | 3.2205 | 17600 | 0.3048 |
|
| 146 |
+
| 0.1354 | 3.2296 | 17650 | 0.3033 |
|
| 147 |
+
| 0.1233 | 3.2388 | 17700 | 0.3001 |
|
| 148 |
+
| 0.1139 | 3.2479 | 17750 | 0.3028 |
|
| 149 |
+
| 0.1297 | 3.2571 | 17800 | 0.3026 |
|
| 150 |
+
| 0.1339 | 3.2662 | 17850 | 0.3014 |
|
| 151 |
+
| 0.1244 | 3.2754 | 17900 | 0.3041 |
|
| 152 |
+
| 0.1351 | 3.2845 | 17950 | 0.3051 |
|
| 153 |
+
| 0.1147 | 3.2937 | 18000 | 0.3036 |
|
| 154 |
+
| 0.1251 | 3.3028 | 18050 | 0.3031 |
|
| 155 |
+
| 0.1158 | 3.3120 | 18100 | 0.3030 |
|
| 156 |
+
| 0.1192 | 3.3211 | 18150 | 0.3043 |
|
| 157 |
+
| 0.1293 | 3.3303 | 18200 | 0.3048 |
|
| 158 |
+
| 0.1255 | 3.3394 | 18250 | 0.3006 |
|
| 159 |
+
| 0.1306 | 3.3486 | 18300 | 0.3016 |
|
| 160 |
+
| 0.1195 | 3.3577 | 18350 | 0.3029 |
|
| 161 |
+
| 0.1273 | 3.3669 | 18400 | 0.3043 |
|
| 162 |
+
| 0.1127 | 3.3760 | 18450 | 0.3044 |
|
| 163 |
+
| 0.1315 | 3.3852 | 18500 | 0.3040 |
|
| 164 |
+
| 0.1138 | 3.3943 | 18550 | 0.3018 |
|
| 165 |
+
| 0.1277 | 3.4035 | 18600 | 0.3039 |
|
| 166 |
+
| 0.1139 | 3.4126 | 18650 | 0.3031 |
|
| 167 |
+
| 0.1244 | 3.4218 | 18700 | 0.3019 |
|
| 168 |
+
| 0.1323 | 3.4309 | 18750 | 0.3013 |
|
| 169 |
+
| 0.128 | 3.4401 | 18800 | 0.3015 |
|
| 170 |
+
| 0.1232 | 3.4492 | 18850 | 0.3035 |
|
| 171 |
+
| 0.1276 | 3.4584 | 18900 | 0.3032 |
|
| 172 |
+
| 0.1258 | 3.4675 | 18950 | 0.3037 |
|
| 173 |
+
| 0.1376 | 3.4767 | 19000 | 0.3022 |
|
| 174 |
+
| 0.1337 | 3.4858 | 19050 | 0.3019 |
|
| 175 |
+
| 0.1146 | 3.4950 | 19100 | 0.3028 |
|
| 176 |
+
| 0.115 | 3.5041 | 19150 | 0.3020 |
|
| 177 |
+
| 0.1229 | 3.5133 | 19200 | 0.3014 |
|
| 178 |
+
| 0.1324 | 3.5224 | 19250 | 0.3015 |
|
| 179 |
+
| 0.1294 | 3.5316 | 19300 | 0.3026 |
|
| 180 |
+
| 0.1284 | 3.5407 | 19350 | 0.3004 |
|
| 181 |
+
| 0.1247 | 3.5499 | 19400 | 0.3029 |
|
| 182 |
+
| 0.1227 | 3.5590 | 19450 | 0.3018 |
|
| 183 |
+
| 0.1192 | 3.5682 | 19500 | 0.3007 |
|
| 184 |
+
| 0.1204 | 3.5773 | 19550 | 0.3021 |
|
| 185 |
+
| 0.1124 | 3.5865 | 19600 | 0.3035 |
|
| 186 |
+
| 0.1254 | 3.5956 | 19650 | 0.3028 |
|
| 187 |
+
| 0.1225 | 3.6048 | 19700 | 0.3018 |
|
| 188 |
+
| 0.1198 | 3.6139 | 19750 | 0.3027 |
|
| 189 |
+
| 0.1245 | 3.6231 | 19800 | 0.3023 |
|
| 190 |
+
| 0.1175 | 3.6322 | 19850 | 0.3031 |
|
| 191 |
+
| 0.113 | 3.6414 | 19900 | 0.3034 |
|
| 192 |
+
| 0.1241 | 3.6505 | 19950 | 0.3028 |
|
| 193 |
+
| 0.1333 | 3.6597 | 20000 | 0.3028 |
|
| 194 |
+
| 0.1159 | 3.6688 | 20050 | 0.3016 |
|
| 195 |
+
| 0.1208 | 3.6780 | 20100 | 0.3015 |
|
| 196 |
+
| 0.1327 | 3.6871 | 20150 | 0.3002 |
|
| 197 |
+
| 0.1176 | 3.6962 | 20200 | 0.3020 |
|
| 198 |
+
| 0.1149 | 3.7054 | 20250 | 0.3020 |
|
| 199 |
+
| 0.1126 | 3.7145 | 20300 | 0.3013 |
|
| 200 |
+
| 0.1238 | 3.7237 | 20350 | 0.3001 |
|
| 201 |
+
| 0.1129 | 3.7328 | 20400 | 0.3006 |
|
| 202 |
+
| 0.1259 | 3.7420 | 20450 | 0.3004 |
|
| 203 |
+
| 0.1186 | 3.7511 | 20500 | 0.3004 |
|
| 204 |
+
| 0.1257 | 3.7603 | 20550 | 0.3006 |
|
| 205 |
+
| 0.1265 | 3.7694 | 20600 | 0.3006 |
|
| 206 |
+
| 0.1257 | 3.7786 | 20650 | 0.3005 |
|
| 207 |
+
| 0.1167 | 3.7877 | 20700 | 0.3000 |
|
| 208 |
+
| 0.1121 | 3.7969 | 20750 | 0.3010 |
|
| 209 |
+
| 0.1201 | 3.8060 | 20800 | 0.3002 |
|
| 210 |
+
| 0.1299 | 3.8152 | 20850 | 0.3004 |
|
| 211 |
+
| 0.1325 | 3.8243 | 20900 | 0.2993 |
|
| 212 |
+
| 0.1221 | 3.8335 | 20950 | 0.2999 |
|
| 213 |
+
| 0.1166 | 3.8426 | 21000 | 0.3004 |
|
| 214 |
+
| 0.1149 | 3.8518 | 21050 | 0.3009 |
|
| 215 |
+
| 0.1094 | 3.8609 | 21100 | 0.3007 |
|
| 216 |
+
| 0.1267 | 3.8701 | 21150 | 0.2994 |
|
| 217 |
+
| 0.1246 | 3.8792 | 21200 | 0.2995 |
|
| 218 |
+
| 0.1217 | 3.8884 | 21250 | 0.2994 |
|
| 219 |
+
| 0.1133 | 3.8975 | 21300 | 0.2993 |
|
| 220 |
+
| 0.1106 | 3.9067 | 21350 | 0.2998 |
|
| 221 |
+
| 0.116 | 3.9158 | 21400 | 0.3001 |
|
| 222 |
+
| 0.1155 | 3.9250 | 21450 | 0.2998 |
|
| 223 |
+
| 0.1249 | 3.9341 | 21500 | 0.3002 |
|
| 224 |
+
| 0.1203 | 3.9433 | 21550 | 0.3003 |
|
| 225 |
+
| 0.1307 | 3.9524 | 21600 | 0.3001 |
|
| 226 |
+
| 0.1328 | 3.9616 | 21650 | 0.3001 |
|
| 227 |
+
| 0.1202 | 3.9707 | 21700 | 0.3001 |
|
| 228 |
+
| 0.1283 | 3.9799 | 21750 | 0.3001 |
|
| 229 |
+
| 0.1133 | 3.9890 | 21800 | 0.3001 |
|
| 230 |
+
| 0.1198 | 3.9982 | 21850 | 0.3001 |
|
| 231 |
|
| 232 |
|
| 233 |
### Framework versions
|
adapter_config.json
CHANGED
|
@@ -20,13 +20,13 @@
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
| 23 |
-
"q_proj",
|
| 24 |
-
"o_proj",
|
| 25 |
-
"up_proj",
|
| 26 |
"down_proj",
|
|
|
|
|
|
|
|
|
|
| 27 |
"k_proj",
|
| 28 |
-
"
|
| 29 |
-
"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
|
|
|
| 20 |
"rank_pattern": {},
|
| 21 |
"revision": null,
|
| 22 |
"target_modules": [
|
|
|
|
|
|
|
|
|
|
| 23 |
"down_proj",
|
| 24 |
+
"v_proj",
|
| 25 |
+
"o_proj",
|
| 26 |
+
"q_proj",
|
| 27 |
"k_proj",
|
| 28 |
+
"up_proj",
|
| 29 |
+
"gate_proj"
|
| 30 |
],
|
| 31 |
"task_type": "CAUSAL_LM",
|
| 32 |
"use_dora": false,
|
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2332095256
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:238b4089776c079e290390cae696f344fbbb4d6d76d11bbab09c7beee67abb7d
|
| 3 |
size 2332095256
|
runs/Sep10_07-54-22_f5f5cac154cc/events.out.tfevents.1725954874.f5f5cac154cc.4087.0
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1b9964f61f1e5199ac5dd5785ffa61bb67d434f6f52d999fb1e53a98f0eebbfd
|
| 3 |
+
size 147657
|
training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5496
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0733919e084b1dbf67005aa752101fd155eb6e4673984ff1875538c3de9dc07e
|
| 3 |
size 5496
|