| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.0, | |
| "eval_steps": 12, | |
| "global_step": 47, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.02127659574468085, | |
| "grad_norm": 0.42955278162899063, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.1272, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.02127659574468085, | |
| "eval_loss": 1.130263328552246, | |
| "eval_runtime": 46.1353, | |
| "eval_samples_per_second": 2.146, | |
| "eval_steps_per_second": 0.217, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0425531914893617, | |
| "grad_norm": 0.35905455037073136, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.2237, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.06382978723404255, | |
| "grad_norm": 0.3859662975436329, | |
| "learning_rate": 3e-06, | |
| "loss": 1.2109, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0851063829787234, | |
| "grad_norm": 0.3458897777003975, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.3168, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.10638297872340426, | |
| "grad_norm": 0.297024954598104, | |
| "learning_rate": 5e-06, | |
| "loss": 1.1879, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.1276595744680851, | |
| "grad_norm": 0.2974375637230888, | |
| "learning_rate": 6e-06, | |
| "loss": 1.1898, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14893617021276595, | |
| "grad_norm": 0.35483433691691424, | |
| "learning_rate": 7e-06, | |
| "loss": 1.2275, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.1702127659574468, | |
| "grad_norm": 0.3817564026852017, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2163, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.19148936170212766, | |
| "grad_norm": 0.40183241982975176, | |
| "learning_rate": 9e-06, | |
| "loss": 1.0668, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2127659574468085, | |
| "grad_norm": 0.3748400230734179, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2589, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.23404255319148937, | |
| "grad_norm": 0.4732373123638865, | |
| "learning_rate": 9.981987442712634e-06, | |
| "loss": 1.209, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "grad_norm": 0.34097014836642403, | |
| "learning_rate": 9.928079551738542e-06, | |
| "loss": 1.1222, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2553191489361702, | |
| "eval_loss": 1.1231276988983154, | |
| "eval_runtime": 45.5524, | |
| "eval_samples_per_second": 2.173, | |
| "eval_steps_per_second": 0.22, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.2765957446808511, | |
| "grad_norm": 0.3865545405035091, | |
| "learning_rate": 9.838664734667496e-06, | |
| "loss": 1.2509, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.2978723404255319, | |
| "grad_norm": 0.5747305954760134, | |
| "learning_rate": 9.714387227305422e-06, | |
| "loss": 1.1448, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.3191489361702128, | |
| "grad_norm": 0.43342726982955576, | |
| "learning_rate": 9.55614245194068e-06, | |
| "loss": 1.1433, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.3404255319148936, | |
| "grad_norm": 0.38621360188314374, | |
| "learning_rate": 9.365070565805941e-06, | |
| "loss": 1.2438, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.3617021276595745, | |
| "grad_norm": 0.4586858888503175, | |
| "learning_rate": 9.142548246219212e-06, | |
| "loss": 1.217, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.3829787234042553, | |
| "grad_norm": 0.4605021995799594, | |
| "learning_rate": 8.890178771592198e-06, | |
| "loss": 1.1023, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.40425531914893614, | |
| "grad_norm": 0.4149444720869162, | |
| "learning_rate": 8.609780469772623e-06, | |
| "loss": 1.1057, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.425531914893617, | |
| "grad_norm": 0.4776677990058847, | |
| "learning_rate": 8.303373616950408e-06, | |
| "loss": 1.1878, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.44680851063829785, | |
| "grad_norm": 0.6083128734022943, | |
| "learning_rate": 7.973165881521435e-06, | |
| "loss": 1.0719, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.46808510638297873, | |
| "grad_norm": 0.427407991852223, | |
| "learning_rate": 7.621536417786159e-06, | |
| "loss": 1.1414, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.48936170212765956, | |
| "grad_norm": 0.41689291665199874, | |
| "learning_rate": 7.251018724088367e-06, | |
| "loss": 1.1651, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "grad_norm": 0.5049216829996077, | |
| "learning_rate": 6.864282388901544e-06, | |
| "loss": 1.1099, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5106382978723404, | |
| "eval_loss": 1.0642483234405518, | |
| "eval_runtime": 45.5272, | |
| "eval_samples_per_second": 2.175, | |
| "eval_steps_per_second": 0.22, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.5319148936170213, | |
| "grad_norm": 0.3940106585784572, | |
| "learning_rate": 6.464113856382752e-06, | |
| "loss": 0.9954, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.5531914893617021, | |
| "grad_norm": 0.35275642793537804, | |
| "learning_rate": 6.053396349978632e-06, | |
| "loss": 1.0469, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.574468085106383, | |
| "grad_norm": 0.5184071159973599, | |
| "learning_rate": 5.635089098734394e-06, | |
| "loss": 1.1481, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.5957446808510638, | |
| "grad_norm": 0.4370219832674898, | |
| "learning_rate": 5.212206015980742e-06, | |
| "loss": 1.1364, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.6170212765957447, | |
| "grad_norm": 0.36478900402279923, | |
| "learning_rate": 4.78779398401926e-06, | |
| "loss": 1.0762, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.6382978723404256, | |
| "grad_norm": 0.40328712656577087, | |
| "learning_rate": 4.364910901265607e-06, | |
| "loss": 1.085, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.6595744680851063, | |
| "grad_norm": 0.341826695457334, | |
| "learning_rate": 3.94660365002137e-06, | |
| "loss": 1.0013, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.6808510638297872, | |
| "grad_norm": 0.4177223934587119, | |
| "learning_rate": 3.5358861436172487e-06, | |
| "loss": 1.0372, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.7021276595744681, | |
| "grad_norm": 0.4218122745580511, | |
| "learning_rate": 3.1357176110984578e-06, | |
| "loss": 0.9904, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.723404255319149, | |
| "grad_norm": 0.3778418743271345, | |
| "learning_rate": 2.748981275911633e-06, | |
| "loss": 1.0609, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.7446808510638298, | |
| "grad_norm": 0.5024526255778048, | |
| "learning_rate": 2.3784635822138424e-06, | |
| "loss": 1.0193, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "grad_norm": 0.29310915204897864, | |
| "learning_rate": 2.0268341184785674e-06, | |
| "loss": 1.105, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7659574468085106, | |
| "eval_loss": 1.0241957902908325, | |
| "eval_runtime": 45.4699, | |
| "eval_samples_per_second": 2.177, | |
| "eval_steps_per_second": 0.22, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.7872340425531915, | |
| "grad_norm": 0.38019500098636755, | |
| "learning_rate": 1.6966263830495939e-06, | |
| "loss": 1.122, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.8085106382978723, | |
| "grad_norm": 0.4515729624152646, | |
| "learning_rate": 1.390219530227378e-06, | |
| "loss": 1.1268, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.8297872340425532, | |
| "grad_norm": 0.3174904808093991, | |
| "learning_rate": 1.1098212284078037e-06, | |
| "loss": 1.1366, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.851063829787234, | |
| "grad_norm": 0.32824416887732244, | |
| "learning_rate": 8.574517537807897e-07, | |
| "loss": 1.085, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8723404255319149, | |
| "grad_norm": 0.3723789314131866, | |
| "learning_rate": 6.349294341940593e-07, | |
| "loss": 1.0074, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.8936170212765957, | |
| "grad_norm": 0.3863665218570595, | |
| "learning_rate": 4.43857548059321e-07, | |
| "loss": 1.0786, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.9148936170212766, | |
| "grad_norm": 0.28822889001191754, | |
| "learning_rate": 2.85612772694579e-07, | |
| "loss": 1.0286, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.9361702127659575, | |
| "grad_norm": 0.4310630390723365, | |
| "learning_rate": 1.6133526533250566e-07, | |
| "loss": 0.8951, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.9574468085106383, | |
| "grad_norm": 0.3151119202195982, | |
| "learning_rate": 7.192044826145772e-08, | |
| "loss": 0.9489, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.9787234042553191, | |
| "grad_norm": 0.3577888537210098, | |
| "learning_rate": 1.8012557287367394e-08, | |
| "loss": 1.1146, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.383122905902664, | |
| "learning_rate": 0.0, | |
| "loss": 1.031, | |
| "step": 47 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 47, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 286172412641280.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |