| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1272264631043257, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.002544529262086514, | |
| "grad_norm": 3.013115882873535, | |
| "learning_rate": 5e-06, | |
| "loss": 1.1819, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002544529262086514, | |
| "eval_loss": 1.6131161451339722, | |
| "eval_runtime": 4.1535, | |
| "eval_samples_per_second": 39.966, | |
| "eval_steps_per_second": 19.983, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.005089058524173028, | |
| "grad_norm": 3.4997785091400146, | |
| "learning_rate": 1e-05, | |
| "loss": 1.447, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.007633587786259542, | |
| "grad_norm": 2.859135627746582, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.1831, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010178117048346057, | |
| "grad_norm": 2.542371988296509, | |
| "learning_rate": 2e-05, | |
| "loss": 0.8844, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.01272264631043257, | |
| "grad_norm": 2.9563982486724854, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.2164, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.015267175572519083, | |
| "grad_norm": 3.415498733520508, | |
| "learning_rate": 3e-05, | |
| "loss": 1.3261, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.017811704834605598, | |
| "grad_norm": 3.1173617839813232, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.2456, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.020356234096692113, | |
| "grad_norm": 2.952441453933716, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0949, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.022900763358778626, | |
| "grad_norm": 3.2792415618896484, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1191, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.02544529262086514, | |
| "grad_norm": 2.428950786590576, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9415, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.027989821882951654, | |
| "grad_norm": 3.6672072410583496, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 1.1482, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.030534351145038167, | |
| "grad_norm": 2.3909623622894287, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 0.7157, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03307888040712468, | |
| "grad_norm": 2.3522462844848633, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 0.5238, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.03307888040712468, | |
| "eval_loss": 0.7261966466903687, | |
| "eval_runtime": 4.1767, | |
| "eval_samples_per_second": 39.744, | |
| "eval_steps_per_second": 19.872, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.035623409669211195, | |
| "grad_norm": 3.122359275817871, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 0.7318, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03816793893129771, | |
| "grad_norm": 5.089092254638672, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 0.8987, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.04071246819338423, | |
| "grad_norm": 2.803917407989502, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 0.5268, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.043256997455470736, | |
| "grad_norm": 2.718916654586792, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 0.6317, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04580152671755725, | |
| "grad_norm": 2.230231285095215, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 0.2857, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.04834605597964377, | |
| "grad_norm": 1.5982710123062134, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 0.1697, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05089058524173028, | |
| "grad_norm": 1.402782917022705, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.2827, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05343511450381679, | |
| "grad_norm": 1.7604286670684814, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 0.1834, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.05597964376590331, | |
| "grad_norm": 4.199867248535156, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 0.4652, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.058524173027989825, | |
| "grad_norm": 1.1508660316467285, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 0.1081, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.061068702290076333, | |
| "grad_norm": 1.6623399257659912, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 0.1578, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06361323155216285, | |
| "grad_norm": 1.1933339834213257, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 0.1515, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06615776081424936, | |
| "grad_norm": 1.6038551330566406, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 0.2152, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06615776081424936, | |
| "eval_loss": 0.18286243081092834, | |
| "eval_runtime": 4.0988, | |
| "eval_samples_per_second": 40.499, | |
| "eval_steps_per_second": 20.25, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.06870229007633588, | |
| "grad_norm": 1.0369246006011963, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 0.1151, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07124681933842239, | |
| "grad_norm": 1.37936270236969, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 0.1626, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0737913486005089, | |
| "grad_norm": 0.9308303594589233, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 0.1095, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.07633587786259542, | |
| "grad_norm": 1.8044286966323853, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0947, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.07888040712468193, | |
| "grad_norm": 1.0868867635726929, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 0.2049, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.08142493638676845, | |
| "grad_norm": 2.401944875717163, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 0.2697, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.08396946564885496, | |
| "grad_norm": 1.6872973442077637, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 0.1731, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.08651399491094147, | |
| "grad_norm": 1.331525206565857, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 0.16, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.089058524173028, | |
| "grad_norm": 2.928229570388794, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 0.2078, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0916030534351145, | |
| "grad_norm": 1.3831185102462769, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 0.0663, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.09414758269720101, | |
| "grad_norm": 1.482015609741211, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 0.1134, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.09669211195928754, | |
| "grad_norm": 1.5263253450393677, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 0.1705, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.09923664122137404, | |
| "grad_norm": 1.535754680633545, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 0.1579, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.09923664122137404, | |
| "eval_loss": 0.16544334590435028, | |
| "eval_runtime": 4.1198, | |
| "eval_samples_per_second": 40.293, | |
| "eval_steps_per_second": 20.147, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.10178117048346055, | |
| "grad_norm": 1.3084092140197754, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.1543, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.10432569974554708, | |
| "grad_norm": 2.607229232788086, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 0.2427, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.10687022900763359, | |
| "grad_norm": 2.31429123878479, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 0.1742, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.10941475826972011, | |
| "grad_norm": 2.2048234939575195, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 0.1644, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.11195928753180662, | |
| "grad_norm": 0.9953597784042358, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 0.0656, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.11450381679389313, | |
| "grad_norm": 1.6188597679138184, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 0.155, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.11704834605597965, | |
| "grad_norm": 1.3625197410583496, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 0.1116, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.11959287531806616, | |
| "grad_norm": 0.8615650534629822, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 0.038, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.12213740458015267, | |
| "grad_norm": 1.747086763381958, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 0.3618, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.12468193384223919, | |
| "grad_norm": 1.5428842306137085, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 0.1127, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.1272264631043257, | |
| "grad_norm": 1.6774532794952393, | |
| "learning_rate": 0.0, | |
| "loss": 0.1539, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 13, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2500486653542400.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |