Upload folder using huggingface_hub
Browse files- adapter_model.safetensors +1 -1
- optimizer.pt +1 -1
- scheduler.pt +1 -1
- trainer_state.json +703 -3
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 2123042352
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8fb49ede251228dcb0b8261efcdfa84f3f864a3e6875cdd344469187e6108959
|
| 3 |
size 2123042352
|
optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 4246317377
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e51fed16ec147ddedc59a297f257364173675fec3263af1f9737562ea8acee0a
|
| 3 |
size 4246317377
|
scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:992403723697967924269ce9a55a5e4d4fece1b138effdd700fb828d1e9c617e
|
| 3 |
size 1465
|
trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 500,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -2108,6 +2108,706 @@
|
|
| 2108 |
"learning_rate": 4.3230276223144975e-05,
|
| 2109 |
"loss": 4.5361,
|
| 2110 |
"step": 15000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 2111 |
}
|
| 2112 |
],
|
| 2113 |
"logging_steps": 50,
|
|
@@ -2127,7 +2827,7 @@
|
|
| 2127 |
"attributes": {}
|
| 2128 |
}
|
| 2129 |
},
|
| 2130 |
-
"total_flos": 1.
|
| 2131 |
"train_batch_size": 1,
|
| 2132 |
"trial_name": null,
|
| 2133 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.18053800324968405,
|
| 6 |
"eval_steps": 500,
|
| 7 |
+
"global_step": 20000,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 2108 |
"learning_rate": 4.3230276223144975e-05,
|
| 2109 |
"loss": 4.5361,
|
| 2110 |
"step": 15000
|
| 2111 |
+
},
|
| 2112 |
+
{
|
| 2113 |
+
"epoch": 0.13585484744538726,
|
| 2114 |
+
"grad_norm": 3.28633975982666,
|
| 2115 |
+
"learning_rate": 4.320770897273877e-05,
|
| 2116 |
+
"loss": 4.5367,
|
| 2117 |
+
"step": 15050
|
| 2118 |
+
},
|
| 2119 |
+
{
|
| 2120 |
+
"epoch": 0.13630619245351147,
|
| 2121 |
+
"grad_norm": 3.2945947647094727,
|
| 2122 |
+
"learning_rate": 4.318514172233255e-05,
|
| 2123 |
+
"loss": 4.5113,
|
| 2124 |
+
"step": 15100
|
| 2125 |
+
},
|
| 2126 |
+
{
|
| 2127 |
+
"epoch": 0.1367575374616357,
|
| 2128 |
+
"grad_norm": 5.111336708068848,
|
| 2129 |
+
"learning_rate": 4.3162574471926346e-05,
|
| 2130 |
+
"loss": 4.5238,
|
| 2131 |
+
"step": 15150
|
| 2132 |
+
},
|
| 2133 |
+
{
|
| 2134 |
+
"epoch": 0.13720888246975987,
|
| 2135 |
+
"grad_norm": 2.328876256942749,
|
| 2136 |
+
"learning_rate": 4.314000722152013e-05,
|
| 2137 |
+
"loss": 4.5512,
|
| 2138 |
+
"step": 15200
|
| 2139 |
+
},
|
| 2140 |
+
{
|
| 2141 |
+
"epoch": 0.1376602274778841,
|
| 2142 |
+
"grad_norm": 3.703890323638916,
|
| 2143 |
+
"learning_rate": 4.311743997111392e-05,
|
| 2144 |
+
"loss": 4.5192,
|
| 2145 |
+
"step": 15250
|
| 2146 |
+
},
|
| 2147 |
+
{
|
| 2148 |
+
"epoch": 0.1381115724860083,
|
| 2149 |
+
"grad_norm": 2.8396573066711426,
|
| 2150 |
+
"learning_rate": 4.309487272070771e-05,
|
| 2151 |
+
"loss": 4.5641,
|
| 2152 |
+
"step": 15300
|
| 2153 |
+
},
|
| 2154 |
+
{
|
| 2155 |
+
"epoch": 0.13856291749413252,
|
| 2156 |
+
"grad_norm": 3.3222029209136963,
|
| 2157 |
+
"learning_rate": 4.3072305470301496e-05,
|
| 2158 |
+
"loss": 4.5332,
|
| 2159 |
+
"step": 15350
|
| 2160 |
+
},
|
| 2161 |
+
{
|
| 2162 |
+
"epoch": 0.13901426250225674,
|
| 2163 |
+
"grad_norm": 3.652606725692749,
|
| 2164 |
+
"learning_rate": 4.304973821989529e-05,
|
| 2165 |
+
"loss": 4.5056,
|
| 2166 |
+
"step": 15400
|
| 2167 |
+
},
|
| 2168 |
+
{
|
| 2169 |
+
"epoch": 0.13946560751038092,
|
| 2170 |
+
"grad_norm": 7.847742080688477,
|
| 2171 |
+
"learning_rate": 4.3027170969489075e-05,
|
| 2172 |
+
"loss": 4.4884,
|
| 2173 |
+
"step": 15450
|
| 2174 |
+
},
|
| 2175 |
+
{
|
| 2176 |
+
"epoch": 0.13991695251850514,
|
| 2177 |
+
"grad_norm": 3.6494662761688232,
|
| 2178 |
+
"learning_rate": 4.300460371908287e-05,
|
| 2179 |
+
"loss": 4.4938,
|
| 2180 |
+
"step": 15500
|
| 2181 |
+
},
|
| 2182 |
+
{
|
| 2183 |
+
"epoch": 0.14036829752662935,
|
| 2184 |
+
"grad_norm": 4.544933795928955,
|
| 2185 |
+
"learning_rate": 4.298203646867665e-05,
|
| 2186 |
+
"loss": 4.5045,
|
| 2187 |
+
"step": 15550
|
| 2188 |
+
},
|
| 2189 |
+
{
|
| 2190 |
+
"epoch": 0.14081964253475357,
|
| 2191 |
+
"grad_norm": 3.429764986038208,
|
| 2192 |
+
"learning_rate": 4.2959469218270446e-05,
|
| 2193 |
+
"loss": 4.5033,
|
| 2194 |
+
"step": 15600
|
| 2195 |
+
},
|
| 2196 |
+
{
|
| 2197 |
+
"epoch": 0.14127098754287779,
|
| 2198 |
+
"grad_norm": 3.790017604827881,
|
| 2199 |
+
"learning_rate": 4.293690196786423e-05,
|
| 2200 |
+
"loss": 4.4775,
|
| 2201 |
+
"step": 15650
|
| 2202 |
+
},
|
| 2203 |
+
{
|
| 2204 |
+
"epoch": 0.14172233255100197,
|
| 2205 |
+
"grad_norm": 3.4987452030181885,
|
| 2206 |
+
"learning_rate": 4.2914334717458024e-05,
|
| 2207 |
+
"loss": 4.5988,
|
| 2208 |
+
"step": 15700
|
| 2209 |
+
},
|
| 2210 |
+
{
|
| 2211 |
+
"epoch": 0.1421736775591262,
|
| 2212 |
+
"grad_norm": 2.5895438194274902,
|
| 2213 |
+
"learning_rate": 4.289176746705182e-05,
|
| 2214 |
+
"loss": 4.5285,
|
| 2215 |
+
"step": 15750
|
| 2216 |
+
},
|
| 2217 |
+
{
|
| 2218 |
+
"epoch": 0.1426250225672504,
|
| 2219 |
+
"grad_norm": 4.709017276763916,
|
| 2220 |
+
"learning_rate": 4.28692002166456e-05,
|
| 2221 |
+
"loss": 4.4137,
|
| 2222 |
+
"step": 15800
|
| 2223 |
+
},
|
| 2224 |
+
{
|
| 2225 |
+
"epoch": 0.14307636757537462,
|
| 2226 |
+
"grad_norm": 3.9760525226593018,
|
| 2227 |
+
"learning_rate": 4.2846632966239395e-05,
|
| 2228 |
+
"loss": 4.5278,
|
| 2229 |
+
"step": 15850
|
| 2230 |
+
},
|
| 2231 |
+
{
|
| 2232 |
+
"epoch": 0.14352771258349883,
|
| 2233 |
+
"grad_norm": 4.445188045501709,
|
| 2234 |
+
"learning_rate": 4.282406571583318e-05,
|
| 2235 |
+
"loss": 4.5362,
|
| 2236 |
+
"step": 15900
|
| 2237 |
+
},
|
| 2238 |
+
{
|
| 2239 |
+
"epoch": 0.14397905759162305,
|
| 2240 |
+
"grad_norm": 4.021897792816162,
|
| 2241 |
+
"learning_rate": 4.2801498465426974e-05,
|
| 2242 |
+
"loss": 4.458,
|
| 2243 |
+
"step": 15950
|
| 2244 |
+
},
|
| 2245 |
+
{
|
| 2246 |
+
"epoch": 0.14443040259974724,
|
| 2247 |
+
"grad_norm": 4.263660907745361,
|
| 2248 |
+
"learning_rate": 4.277893121502076e-05,
|
| 2249 |
+
"loss": 4.4782,
|
| 2250 |
+
"step": 16000
|
| 2251 |
+
},
|
| 2252 |
+
{
|
| 2253 |
+
"epoch": 0.14488174760787145,
|
| 2254 |
+
"grad_norm": 3.184115171432495,
|
| 2255 |
+
"learning_rate": 4.275636396461455e-05,
|
| 2256 |
+
"loss": 4.4877,
|
| 2257 |
+
"step": 16050
|
| 2258 |
+
},
|
| 2259 |
+
{
|
| 2260 |
+
"epoch": 0.14533309261599567,
|
| 2261 |
+
"grad_norm": 3.6419224739074707,
|
| 2262 |
+
"learning_rate": 4.273379671420834e-05,
|
| 2263 |
+
"loss": 4.5329,
|
| 2264 |
+
"step": 16100
|
| 2265 |
+
},
|
| 2266 |
+
{
|
| 2267 |
+
"epoch": 0.14578443762411988,
|
| 2268 |
+
"grad_norm": 5.209333896636963,
|
| 2269 |
+
"learning_rate": 4.271122946380213e-05,
|
| 2270 |
+
"loss": 4.5252,
|
| 2271 |
+
"step": 16150
|
| 2272 |
+
},
|
| 2273 |
+
{
|
| 2274 |
+
"epoch": 0.1462357826322441,
|
| 2275 |
+
"grad_norm": 2.9980499744415283,
|
| 2276 |
+
"learning_rate": 4.2688662213395924e-05,
|
| 2277 |
+
"loss": 4.4491,
|
| 2278 |
+
"step": 16200
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 0.1466871276403683,
|
| 2282 |
+
"grad_norm": 2.8836166858673096,
|
| 2283 |
+
"learning_rate": 4.266609496298971e-05,
|
| 2284 |
+
"loss": 4.524,
|
| 2285 |
+
"step": 16250
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 0.1471384726484925,
|
| 2289 |
+
"grad_norm": 3.24406099319458,
|
| 2290 |
+
"learning_rate": 4.26435277125835e-05,
|
| 2291 |
+
"loss": 4.5629,
|
| 2292 |
+
"step": 16300
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 0.14758981765661672,
|
| 2296 |
+
"grad_norm": 3.78409743309021,
|
| 2297 |
+
"learning_rate": 4.262096046217729e-05,
|
| 2298 |
+
"loss": 4.5051,
|
| 2299 |
+
"step": 16350
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 0.14804116266474093,
|
| 2303 |
+
"grad_norm": 3.738863229751587,
|
| 2304 |
+
"learning_rate": 4.259839321177108e-05,
|
| 2305 |
+
"loss": 4.3699,
|
| 2306 |
+
"step": 16400
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 0.14849250767286515,
|
| 2310 |
+
"grad_norm": 3.1949925422668457,
|
| 2311 |
+
"learning_rate": 4.2575825961364867e-05,
|
| 2312 |
+
"loss": 4.4681,
|
| 2313 |
+
"step": 16450
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 0.14894385268098934,
|
| 2317 |
+
"grad_norm": 3.774017810821533,
|
| 2318 |
+
"learning_rate": 4.255325871095866e-05,
|
| 2319 |
+
"loss": 4.4382,
|
| 2320 |
+
"step": 16500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 0.14939519768911355,
|
| 2324 |
+
"grad_norm": 3.903379201889038,
|
| 2325 |
+
"learning_rate": 4.2530691460552445e-05,
|
| 2326 |
+
"loss": 4.4229,
|
| 2327 |
+
"step": 16550
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 0.14984654269723777,
|
| 2331 |
+
"grad_norm": 2.8182575702667236,
|
| 2332 |
+
"learning_rate": 4.250812421014624e-05,
|
| 2333 |
+
"loss": 4.4755,
|
| 2334 |
+
"step": 16600
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 0.15029788770536198,
|
| 2338 |
+
"grad_norm": 3.8375935554504395,
|
| 2339 |
+
"learning_rate": 4.2485556959740024e-05,
|
| 2340 |
+
"loss": 4.5113,
|
| 2341 |
+
"step": 16650
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 0.1507492327134862,
|
| 2345 |
+
"grad_norm": 3.6683831214904785,
|
| 2346 |
+
"learning_rate": 4.2462989709333816e-05,
|
| 2347 |
+
"loss": 4.5386,
|
| 2348 |
+
"step": 16700
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 0.15120057772161039,
|
| 2352 |
+
"grad_norm": 4.0321431159973145,
|
| 2353 |
+
"learning_rate": 4.244042245892761e-05,
|
| 2354 |
+
"loss": 4.4977,
|
| 2355 |
+
"step": 16750
|
| 2356 |
+
},
|
| 2357 |
+
{
|
| 2358 |
+
"epoch": 0.1516519227297346,
|
| 2359 |
+
"grad_norm": 3.8294458389282227,
|
| 2360 |
+
"learning_rate": 4.2417855208521395e-05,
|
| 2361 |
+
"loss": 4.4926,
|
| 2362 |
+
"step": 16800
|
| 2363 |
+
},
|
| 2364 |
+
{
|
| 2365 |
+
"epoch": 0.15210326773785882,
|
| 2366 |
+
"grad_norm": 3.6209237575531006,
|
| 2367 |
+
"learning_rate": 4.239528795811519e-05,
|
| 2368 |
+
"loss": 4.5028,
|
| 2369 |
+
"step": 16850
|
| 2370 |
+
},
|
| 2371 |
+
{
|
| 2372 |
+
"epoch": 0.15255461274598303,
|
| 2373 |
+
"grad_norm": 3.8138227462768555,
|
| 2374 |
+
"learning_rate": 4.237272070770897e-05,
|
| 2375 |
+
"loss": 4.4808,
|
| 2376 |
+
"step": 16900
|
| 2377 |
+
},
|
| 2378 |
+
{
|
| 2379 |
+
"epoch": 0.15300595775410725,
|
| 2380 |
+
"grad_norm": 4.5005927085876465,
|
| 2381 |
+
"learning_rate": 4.2350153457302766e-05,
|
| 2382 |
+
"loss": 4.4702,
|
| 2383 |
+
"step": 16950
|
| 2384 |
+
},
|
| 2385 |
+
{
|
| 2386 |
+
"epoch": 0.15345730276223146,
|
| 2387 |
+
"grad_norm": 3.48544979095459,
|
| 2388 |
+
"learning_rate": 4.232758620689655e-05,
|
| 2389 |
+
"loss": 4.4993,
|
| 2390 |
+
"step": 17000
|
| 2391 |
+
},
|
| 2392 |
+
{
|
| 2393 |
+
"epoch": 0.15390864777035565,
|
| 2394 |
+
"grad_norm": 3.5820982456207275,
|
| 2395 |
+
"learning_rate": 4.2305018956490344e-05,
|
| 2396 |
+
"loss": 4.5032,
|
| 2397 |
+
"step": 17050
|
| 2398 |
+
},
|
| 2399 |
+
{
|
| 2400 |
+
"epoch": 0.15435999277847987,
|
| 2401 |
+
"grad_norm": 4.8123555183410645,
|
| 2402 |
+
"learning_rate": 4.228245170608413e-05,
|
| 2403 |
+
"loss": 4.5196,
|
| 2404 |
+
"step": 17100
|
| 2405 |
+
},
|
| 2406 |
+
{
|
| 2407 |
+
"epoch": 0.15481133778660408,
|
| 2408 |
+
"grad_norm": 3.8024814128875732,
|
| 2409 |
+
"learning_rate": 4.225988445567792e-05,
|
| 2410 |
+
"loss": 4.5327,
|
| 2411 |
+
"step": 17150
|
| 2412 |
+
},
|
| 2413 |
+
{
|
| 2414 |
+
"epoch": 0.1552626827947283,
|
| 2415 |
+
"grad_norm": 5.407778263092041,
|
| 2416 |
+
"learning_rate": 4.223731720527171e-05,
|
| 2417 |
+
"loss": 4.5355,
|
| 2418 |
+
"step": 17200
|
| 2419 |
+
},
|
| 2420 |
+
{
|
| 2421 |
+
"epoch": 0.1557140278028525,
|
| 2422 |
+
"grad_norm": 3.6917614936828613,
|
| 2423 |
+
"learning_rate": 4.22147499548655e-05,
|
| 2424 |
+
"loss": 4.4072,
|
| 2425 |
+
"step": 17250
|
| 2426 |
+
},
|
| 2427 |
+
{
|
| 2428 |
+
"epoch": 0.1561653728109767,
|
| 2429 |
+
"grad_norm": 3.9421164989471436,
|
| 2430 |
+
"learning_rate": 4.2192182704459294e-05,
|
| 2431 |
+
"loss": 4.4812,
|
| 2432 |
+
"step": 17300
|
| 2433 |
+
},
|
| 2434 |
+
{
|
| 2435 |
+
"epoch": 0.15661671781910091,
|
| 2436 |
+
"grad_norm": 4.172101974487305,
|
| 2437 |
+
"learning_rate": 4.216961545405308e-05,
|
| 2438 |
+
"loss": 4.4737,
|
| 2439 |
+
"step": 17350
|
| 2440 |
+
},
|
| 2441 |
+
{
|
| 2442 |
+
"epoch": 0.15706806282722513,
|
| 2443 |
+
"grad_norm": 3.308185577392578,
|
| 2444 |
+
"learning_rate": 4.214704820364687e-05,
|
| 2445 |
+
"loss": 4.5789,
|
| 2446 |
+
"step": 17400
|
| 2447 |
+
},
|
| 2448 |
+
{
|
| 2449 |
+
"epoch": 0.15751940783534935,
|
| 2450 |
+
"grad_norm": 4.956492900848389,
|
| 2451 |
+
"learning_rate": 4.212448095324066e-05,
|
| 2452 |
+
"loss": 4.565,
|
| 2453 |
+
"step": 17450
|
| 2454 |
+
},
|
| 2455 |
+
{
|
| 2456 |
+
"epoch": 0.15797075284347356,
|
| 2457 |
+
"grad_norm": 3.411794900894165,
|
| 2458 |
+
"learning_rate": 4.210191370283445e-05,
|
| 2459 |
+
"loss": 4.5473,
|
| 2460 |
+
"step": 17500
|
| 2461 |
+
},
|
| 2462 |
+
{
|
| 2463 |
+
"epoch": 0.15842209785159775,
|
| 2464 |
+
"grad_norm": 4.067993640899658,
|
| 2465 |
+
"learning_rate": 4.207934645242824e-05,
|
| 2466 |
+
"loss": 4.4836,
|
| 2467 |
+
"step": 17550
|
| 2468 |
+
},
|
| 2469 |
+
{
|
| 2470 |
+
"epoch": 0.15887344285972196,
|
| 2471 |
+
"grad_norm": 2.9520280361175537,
|
| 2472 |
+
"learning_rate": 4.205677920202203e-05,
|
| 2473 |
+
"loss": 4.4962,
|
| 2474 |
+
"step": 17600
|
| 2475 |
+
},
|
| 2476 |
+
{
|
| 2477 |
+
"epoch": 0.15932478786784618,
|
| 2478 |
+
"grad_norm": 4.387596130371094,
|
| 2479 |
+
"learning_rate": 4.2034211951615815e-05,
|
| 2480 |
+
"loss": 4.513,
|
| 2481 |
+
"step": 17650
|
| 2482 |
+
},
|
| 2483 |
+
{
|
| 2484 |
+
"epoch": 0.1597761328759704,
|
| 2485 |
+
"grad_norm": 3.250239849090576,
|
| 2486 |
+
"learning_rate": 4.201164470120961e-05,
|
| 2487 |
+
"loss": 4.5496,
|
| 2488 |
+
"step": 17700
|
| 2489 |
+
},
|
| 2490 |
+
{
|
| 2491 |
+
"epoch": 0.1602274778840946,
|
| 2492 |
+
"grad_norm": 3.867882013320923,
|
| 2493 |
+
"learning_rate": 4.1989077450803394e-05,
|
| 2494 |
+
"loss": 4.5849,
|
| 2495 |
+
"step": 17750
|
| 2496 |
+
},
|
| 2497 |
+
{
|
| 2498 |
+
"epoch": 0.16067882289221883,
|
| 2499 |
+
"grad_norm": 3.7500853538513184,
|
| 2500 |
+
"learning_rate": 4.196651020039719e-05,
|
| 2501 |
+
"loss": 4.4585,
|
| 2502 |
+
"step": 17800
|
| 2503 |
+
},
|
| 2504 |
+
{
|
| 2505 |
+
"epoch": 0.161130167900343,
|
| 2506 |
+
"grad_norm": 3.8945131301879883,
|
| 2507 |
+
"learning_rate": 4.194394294999098e-05,
|
| 2508 |
+
"loss": 4.4149,
|
| 2509 |
+
"step": 17850
|
| 2510 |
+
},
|
| 2511 |
+
{
|
| 2512 |
+
"epoch": 0.16158151290846723,
|
| 2513 |
+
"grad_norm": 8.667535781860352,
|
| 2514 |
+
"learning_rate": 4.1921375699584765e-05,
|
| 2515 |
+
"loss": 4.44,
|
| 2516 |
+
"step": 17900
|
| 2517 |
+
},
|
| 2518 |
+
{
|
| 2519 |
+
"epoch": 0.16203285791659144,
|
| 2520 |
+
"grad_norm": 4.284276485443115,
|
| 2521 |
+
"learning_rate": 4.189880844917856e-05,
|
| 2522 |
+
"loss": 4.4561,
|
| 2523 |
+
"step": 17950
|
| 2524 |
+
},
|
| 2525 |
+
{
|
| 2526 |
+
"epoch": 0.16248420292471566,
|
| 2527 |
+
"grad_norm": 2.9393467903137207,
|
| 2528 |
+
"learning_rate": 4.1876241198772344e-05,
|
| 2529 |
+
"loss": 4.5887,
|
| 2530 |
+
"step": 18000
|
| 2531 |
+
},
|
| 2532 |
+
{
|
| 2533 |
+
"epoch": 0.16293554793283987,
|
| 2534 |
+
"grad_norm": 3.012742519378662,
|
| 2535 |
+
"learning_rate": 4.1853673948366136e-05,
|
| 2536 |
+
"loss": 4.4513,
|
| 2537 |
+
"step": 18050
|
| 2538 |
+
},
|
| 2539 |
+
{
|
| 2540 |
+
"epoch": 0.16338689294096406,
|
| 2541 |
+
"grad_norm": 5.467082500457764,
|
| 2542 |
+
"learning_rate": 4.183110669795992e-05,
|
| 2543 |
+
"loss": 4.5611,
|
| 2544 |
+
"step": 18100
|
| 2545 |
+
},
|
| 2546 |
+
{
|
| 2547 |
+
"epoch": 0.16383823794908828,
|
| 2548 |
+
"grad_norm": 3.46402907371521,
|
| 2549 |
+
"learning_rate": 4.1808539447553715e-05,
|
| 2550 |
+
"loss": 4.5312,
|
| 2551 |
+
"step": 18150
|
| 2552 |
+
},
|
| 2553 |
+
{
|
| 2554 |
+
"epoch": 0.1642895829572125,
|
| 2555 |
+
"grad_norm": 3.8491625785827637,
|
| 2556 |
+
"learning_rate": 4.17859721971475e-05,
|
| 2557 |
+
"loss": 4.4916,
|
| 2558 |
+
"step": 18200
|
| 2559 |
+
},
|
| 2560 |
+
{
|
| 2561 |
+
"epoch": 0.1647409279653367,
|
| 2562 |
+
"grad_norm": 5.8692450523376465,
|
| 2563 |
+
"learning_rate": 4.176340494674129e-05,
|
| 2564 |
+
"loss": 4.4869,
|
| 2565 |
+
"step": 18250
|
| 2566 |
+
},
|
| 2567 |
+
{
|
| 2568 |
+
"epoch": 0.16519227297346092,
|
| 2569 |
+
"grad_norm": 3.2287988662719727,
|
| 2570 |
+
"learning_rate": 4.174083769633508e-05,
|
| 2571 |
+
"loss": 4.4431,
|
| 2572 |
+
"step": 18300
|
| 2573 |
+
},
|
| 2574 |
+
{
|
| 2575 |
+
"epoch": 0.1656436179815851,
|
| 2576 |
+
"grad_norm": 4.350259304046631,
|
| 2577 |
+
"learning_rate": 4.171827044592887e-05,
|
| 2578 |
+
"loss": 4.4968,
|
| 2579 |
+
"step": 18350
|
| 2580 |
+
},
|
| 2581 |
+
{
|
| 2582 |
+
"epoch": 0.16609496298970933,
|
| 2583 |
+
"grad_norm": 3.7243659496307373,
|
| 2584 |
+
"learning_rate": 4.1695703195522664e-05,
|
| 2585 |
+
"loss": 4.4738,
|
| 2586 |
+
"step": 18400
|
| 2587 |
+
},
|
| 2588 |
+
{
|
| 2589 |
+
"epoch": 0.16654630799783354,
|
| 2590 |
+
"grad_norm": 4.834224224090576,
|
| 2591 |
+
"learning_rate": 4.167313594511645e-05,
|
| 2592 |
+
"loss": 4.5754,
|
| 2593 |
+
"step": 18450
|
| 2594 |
+
},
|
| 2595 |
+
{
|
| 2596 |
+
"epoch": 0.16699765300595776,
|
| 2597 |
+
"grad_norm": 6.014001846313477,
|
| 2598 |
+
"learning_rate": 4.165056869471024e-05,
|
| 2599 |
+
"loss": 4.5449,
|
| 2600 |
+
"step": 18500
|
| 2601 |
+
},
|
| 2602 |
+
{
|
| 2603 |
+
"epoch": 0.16744899801408197,
|
| 2604 |
+
"grad_norm": 3.7950220108032227,
|
| 2605 |
+
"learning_rate": 4.162800144430403e-05,
|
| 2606 |
+
"loss": 4.4007,
|
| 2607 |
+
"step": 18550
|
| 2608 |
+
},
|
| 2609 |
+
{
|
| 2610 |
+
"epoch": 0.16790034302220616,
|
| 2611 |
+
"grad_norm": 4.019992828369141,
|
| 2612 |
+
"learning_rate": 4.160543419389782e-05,
|
| 2613 |
+
"loss": 4.4231,
|
| 2614 |
+
"step": 18600
|
| 2615 |
+
},
|
| 2616 |
+
{
|
| 2617 |
+
"epoch": 0.16835168803033038,
|
| 2618 |
+
"grad_norm": 4.363696575164795,
|
| 2619 |
+
"learning_rate": 4.158286694349161e-05,
|
| 2620 |
+
"loss": 4.4445,
|
| 2621 |
+
"step": 18650
|
| 2622 |
+
},
|
| 2623 |
+
{
|
| 2624 |
+
"epoch": 0.1688030330384546,
|
| 2625 |
+
"grad_norm": 4.168088912963867,
|
| 2626 |
+
"learning_rate": 4.15602996930854e-05,
|
| 2627 |
+
"loss": 4.5064,
|
| 2628 |
+
"step": 18700
|
| 2629 |
+
},
|
| 2630 |
+
{
|
| 2631 |
+
"epoch": 0.1692543780465788,
|
| 2632 |
+
"grad_norm": 3.3574249744415283,
|
| 2633 |
+
"learning_rate": 4.1537732442679186e-05,
|
| 2634 |
+
"loss": 4.5161,
|
| 2635 |
+
"step": 18750
|
| 2636 |
+
},
|
| 2637 |
+
{
|
| 2638 |
+
"epoch": 0.16970572305470302,
|
| 2639 |
+
"grad_norm": 4.255382061004639,
|
| 2640 |
+
"learning_rate": 4.151516519227298e-05,
|
| 2641 |
+
"loss": 4.4809,
|
| 2642 |
+
"step": 18800
|
| 2643 |
+
},
|
| 2644 |
+
{
|
| 2645 |
+
"epoch": 0.17015706806282724,
|
| 2646 |
+
"grad_norm": 3.896949291229248,
|
| 2647 |
+
"learning_rate": 4.1492597941866764e-05,
|
| 2648 |
+
"loss": 4.488,
|
| 2649 |
+
"step": 18850
|
| 2650 |
+
},
|
| 2651 |
+
{
|
| 2652 |
+
"epoch": 0.17060841307095143,
|
| 2653 |
+
"grad_norm": 4.572742938995361,
|
| 2654 |
+
"learning_rate": 4.147003069146056e-05,
|
| 2655 |
+
"loss": 4.5692,
|
| 2656 |
+
"step": 18900
|
| 2657 |
+
},
|
| 2658 |
+
{
|
| 2659 |
+
"epoch": 0.17105975807907564,
|
| 2660 |
+
"grad_norm": 4.25124454498291,
|
| 2661 |
+
"learning_rate": 4.144746344105434e-05,
|
| 2662 |
+
"loss": 4.5341,
|
| 2663 |
+
"step": 18950
|
| 2664 |
+
},
|
| 2665 |
+
{
|
| 2666 |
+
"epoch": 0.17151110308719986,
|
| 2667 |
+
"grad_norm": 3.2986035346984863,
|
| 2668 |
+
"learning_rate": 4.1424896190648136e-05,
|
| 2669 |
+
"loss": 4.4489,
|
| 2670 |
+
"step": 19000
|
| 2671 |
+
},
|
| 2672 |
+
{
|
| 2673 |
+
"epoch": 0.17196244809532407,
|
| 2674 |
+
"grad_norm": 3.633592367172241,
|
| 2675 |
+
"learning_rate": 4.140232894024192e-05,
|
| 2676 |
+
"loss": 4.5124,
|
| 2677 |
+
"step": 19050
|
| 2678 |
+
},
|
| 2679 |
+
{
|
| 2680 |
+
"epoch": 0.1724137931034483,
|
| 2681 |
+
"grad_norm": 3.3687500953674316,
|
| 2682 |
+
"learning_rate": 4.1379761689835714e-05,
|
| 2683 |
+
"loss": 4.5165,
|
| 2684 |
+
"step": 19100
|
| 2685 |
+
},
|
| 2686 |
+
{
|
| 2687 |
+
"epoch": 0.17286513811157247,
|
| 2688 |
+
"grad_norm": 4.958398342132568,
|
| 2689 |
+
"learning_rate": 4.13571944394295e-05,
|
| 2690 |
+
"loss": 4.5043,
|
| 2691 |
+
"step": 19150
|
| 2692 |
+
},
|
| 2693 |
+
{
|
| 2694 |
+
"epoch": 0.1733164831196967,
|
| 2695 |
+
"grad_norm": 4.127295017242432,
|
| 2696 |
+
"learning_rate": 4.133462718902329e-05,
|
| 2697 |
+
"loss": 4.4783,
|
| 2698 |
+
"step": 19200
|
| 2699 |
+
},
|
| 2700 |
+
{
|
| 2701 |
+
"epoch": 0.1737678281278209,
|
| 2702 |
+
"grad_norm": 3.3556175231933594,
|
| 2703 |
+
"learning_rate": 4.131205993861708e-05,
|
| 2704 |
+
"loss": 4.4185,
|
| 2705 |
+
"step": 19250
|
| 2706 |
+
},
|
| 2707 |
+
{
|
| 2708 |
+
"epoch": 0.17421917313594512,
|
| 2709 |
+
"grad_norm": 4.382410049438477,
|
| 2710 |
+
"learning_rate": 4.128949268821087e-05,
|
| 2711 |
+
"loss": 4.5009,
|
| 2712 |
+
"step": 19300
|
| 2713 |
+
},
|
| 2714 |
+
{
|
| 2715 |
+
"epoch": 0.17467051814406934,
|
| 2716 |
+
"grad_norm": 3.7760777473449707,
|
| 2717 |
+
"learning_rate": 4.126692543780466e-05,
|
| 2718 |
+
"loss": 4.3572,
|
| 2719 |
+
"step": 19350
|
| 2720 |
+
},
|
| 2721 |
+
{
|
| 2722 |
+
"epoch": 0.17512186315219352,
|
| 2723 |
+
"grad_norm": 4.594768524169922,
|
| 2724 |
+
"learning_rate": 4.124435818739845e-05,
|
| 2725 |
+
"loss": 4.4793,
|
| 2726 |
+
"step": 19400
|
| 2727 |
+
},
|
| 2728 |
+
{
|
| 2729 |
+
"epoch": 0.17557320816031774,
|
| 2730 |
+
"grad_norm": 4.605646133422852,
|
| 2731 |
+
"learning_rate": 4.1221790936992235e-05,
|
| 2732 |
+
"loss": 4.4462,
|
| 2733 |
+
"step": 19450
|
| 2734 |
+
},
|
| 2735 |
+
{
|
| 2736 |
+
"epoch": 0.17602455316844196,
|
| 2737 |
+
"grad_norm": 3.358002185821533,
|
| 2738 |
+
"learning_rate": 4.119922368658603e-05,
|
| 2739 |
+
"loss": 4.4986,
|
| 2740 |
+
"step": 19500
|
| 2741 |
+
},
|
| 2742 |
+
{
|
| 2743 |
+
"epoch": 0.17647589817656617,
|
| 2744 |
+
"grad_norm": 3.7644100189208984,
|
| 2745 |
+
"learning_rate": 4.1176656436179814e-05,
|
| 2746 |
+
"loss": 4.5314,
|
| 2747 |
+
"step": 19550
|
| 2748 |
+
},
|
| 2749 |
+
{
|
| 2750 |
+
"epoch": 0.17692724318469039,
|
| 2751 |
+
"grad_norm": 4.109899044036865,
|
| 2752 |
+
"learning_rate": 4.115408918577361e-05,
|
| 2753 |
+
"loss": 4.5382,
|
| 2754 |
+
"step": 19600
|
| 2755 |
+
},
|
| 2756 |
+
{
|
| 2757 |
+
"epoch": 0.1773785881928146,
|
| 2758 |
+
"grad_norm": 8.232100486755371,
|
| 2759 |
+
"learning_rate": 4.113152193536739e-05,
|
| 2760 |
+
"loss": 4.5095,
|
| 2761 |
+
"step": 19650
|
| 2762 |
+
},
|
| 2763 |
+
{
|
| 2764 |
+
"epoch": 0.1778299332009388,
|
| 2765 |
+
"grad_norm": 3.442411422729492,
|
| 2766 |
+
"learning_rate": 4.1108954684961185e-05,
|
| 2767 |
+
"loss": 4.5861,
|
| 2768 |
+
"step": 19700
|
| 2769 |
+
},
|
| 2770 |
+
{
|
| 2771 |
+
"epoch": 0.178281278209063,
|
| 2772 |
+
"grad_norm": 2.404611825942993,
|
| 2773 |
+
"learning_rate": 4.108638743455497e-05,
|
| 2774 |
+
"loss": 4.4563,
|
| 2775 |
+
"step": 19750
|
| 2776 |
+
},
|
| 2777 |
+
{
|
| 2778 |
+
"epoch": 0.17873262321718722,
|
| 2779 |
+
"grad_norm": 3.3895816802978516,
|
| 2780 |
+
"learning_rate": 4.1063820184148764e-05,
|
| 2781 |
+
"loss": 4.4434,
|
| 2782 |
+
"step": 19800
|
| 2783 |
+
},
|
| 2784 |
+
{
|
| 2785 |
+
"epoch": 0.17918396822531144,
|
| 2786 |
+
"grad_norm": 2.9194042682647705,
|
| 2787 |
+
"learning_rate": 4.104125293374255e-05,
|
| 2788 |
+
"loss": 4.5463,
|
| 2789 |
+
"step": 19850
|
| 2790 |
+
},
|
| 2791 |
+
{
|
| 2792 |
+
"epoch": 0.17963531323343565,
|
| 2793 |
+
"grad_norm": 2.6337718963623047,
|
| 2794 |
+
"learning_rate": 4.101868568333634e-05,
|
| 2795 |
+
"loss": 4.4246,
|
| 2796 |
+
"step": 19900
|
| 2797 |
+
},
|
| 2798 |
+
{
|
| 2799 |
+
"epoch": 0.18008665824155984,
|
| 2800 |
+
"grad_norm": 5.921742916107178,
|
| 2801 |
+
"learning_rate": 4.0996118432930135e-05,
|
| 2802 |
+
"loss": 4.3955,
|
| 2803 |
+
"step": 19950
|
| 2804 |
+
},
|
| 2805 |
+
{
|
| 2806 |
+
"epoch": 0.18053800324968405,
|
| 2807 |
+
"grad_norm": 3.9008045196533203,
|
| 2808 |
+
"learning_rate": 4.097355118252392e-05,
|
| 2809 |
+
"loss": 4.6028,
|
| 2810 |
+
"step": 20000
|
| 2811 |
}
|
| 2812 |
],
|
| 2813 |
"logging_steps": 50,
|
|
|
|
| 2827 |
"attributes": {}
|
| 2828 |
}
|
| 2829 |
},
|
| 2830 |
+
"total_flos": 1.4772471541951488e+17,
|
| 2831 |
"train_batch_size": 1,
|
| 2832 |
"trial_name": null,
|
| 2833 |
"trial_params": null
|