| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0544069640914037, | |
| "eval_steps": 13, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001088139281828074, | |
| "grad_norm": 0.1708880215883255, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2178, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.001088139281828074, | |
| "eval_loss": 0.2808469533920288, | |
| "eval_runtime": 20.7308, | |
| "eval_samples_per_second": 18.668, | |
| "eval_steps_per_second": 9.358, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.002176278563656148, | |
| "grad_norm": 0.18688273429870605, | |
| "learning_rate": 1e-05, | |
| "loss": 0.2729, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.003264417845484222, | |
| "grad_norm": 0.14812062680721283, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.2367, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.004352557127312296, | |
| "grad_norm": 0.18467950820922852, | |
| "learning_rate": 2e-05, | |
| "loss": 0.2042, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.00544069640914037, | |
| "grad_norm": 0.1507299691438675, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.2751, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.006528835690968444, | |
| "grad_norm": 0.20624135434627533, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2811, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.007616974972796518, | |
| "grad_norm": 0.2024683952331543, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.2443, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.008705114254624592, | |
| "grad_norm": 0.16752073168754578, | |
| "learning_rate": 4e-05, | |
| "loss": 0.268, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.009793253536452665, | |
| "grad_norm": 0.17567092180252075, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.2362, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.01088139281828074, | |
| "grad_norm": 0.15081793069839478, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1754, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011969532100108813, | |
| "grad_norm": 0.24378319084644318, | |
| "learning_rate": 4.99229333433282e-05, | |
| "loss": 0.3136, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.013057671381936888, | |
| "grad_norm": 0.17457719147205353, | |
| "learning_rate": 4.9692208514878444e-05, | |
| "loss": 0.162, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.014145810663764961, | |
| "grad_norm": 0.16278016567230225, | |
| "learning_rate": 4.9309248009941914e-05, | |
| "loss": 0.1994, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.014145810663764961, | |
| "eval_loss": 0.25625717639923096, | |
| "eval_runtime": 20.7372, | |
| "eval_samples_per_second": 18.662, | |
| "eval_steps_per_second": 9.355, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.015233949945593036, | |
| "grad_norm": 0.23434151709079742, | |
| "learning_rate": 4.877641290737884e-05, | |
| "loss": 0.2542, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.01632208922742111, | |
| "grad_norm": 0.20760248601436615, | |
| "learning_rate": 4.8096988312782174e-05, | |
| "loss": 0.2256, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.017410228509249184, | |
| "grad_norm": 0.21011245250701904, | |
| "learning_rate": 4.72751631047092e-05, | |
| "loss": 0.2241, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.018498367791077257, | |
| "grad_norm": 0.18041156232357025, | |
| "learning_rate": 4.6316004108852305e-05, | |
| "loss": 0.1655, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.01958650707290533, | |
| "grad_norm": 0.1776050329208374, | |
| "learning_rate": 4.522542485937369e-05, | |
| "loss": 0.2103, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.020674646354733407, | |
| "grad_norm": 0.16318459808826447, | |
| "learning_rate": 4.401014914000078e-05, | |
| "loss": 0.1372, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.02176278563656148, | |
| "grad_norm": 0.18268632888793945, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 0.1775, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.022850924918389554, | |
| "grad_norm": 0.1805269867181778, | |
| "learning_rate": 4.123620120825459e-05, | |
| "loss": 0.1497, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.023939064200217627, | |
| "grad_norm": 0.2117743194103241, | |
| "learning_rate": 3.969463130731183e-05, | |
| "loss": 0.1659, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.025027203482045703, | |
| "grad_norm": 0.20967699587345123, | |
| "learning_rate": 3.8062464117898724e-05, | |
| "loss": 0.1668, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.026115342763873776, | |
| "grad_norm": 0.23078809678554535, | |
| "learning_rate": 3.634976249348867e-05, | |
| "loss": 0.1963, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.02720348204570185, | |
| "grad_norm": 0.27512913942337036, | |
| "learning_rate": 3.456708580912725e-05, | |
| "loss": 0.1898, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.028291621327529923, | |
| "grad_norm": 0.24129854142665863, | |
| "learning_rate": 3.272542485937369e-05, | |
| "loss": 0.1633, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.028291621327529923, | |
| "eval_loss": 0.21075604856014252, | |
| "eval_runtime": 20.7545, | |
| "eval_samples_per_second": 18.647, | |
| "eval_steps_per_second": 9.347, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.029379760609358, | |
| "grad_norm": 0.19251622259616852, | |
| "learning_rate": 3.083613409639764e-05, | |
| "loss": 0.1362, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.030467899891186073, | |
| "grad_norm": 0.18412667512893677, | |
| "learning_rate": 2.8910861626005776e-05, | |
| "loss": 0.1466, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.031556039173014146, | |
| "grad_norm": 0.28334978222846985, | |
| "learning_rate": 2.6961477393196126e-05, | |
| "loss": 0.2359, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.03264417845484222, | |
| "grad_norm": 0.22527562081813812, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.1213, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03373231773667029, | |
| "grad_norm": 0.31526511907577515, | |
| "learning_rate": 2.303852260680388e-05, | |
| "loss": 0.2167, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.03482045701849837, | |
| "grad_norm": 0.14839228987693787, | |
| "learning_rate": 2.1089138373994223e-05, | |
| "loss": 0.1137, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.035908596300326445, | |
| "grad_norm": 0.208391934633255, | |
| "learning_rate": 1.9163865903602374e-05, | |
| "loss": 0.1615, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.036996735582154515, | |
| "grad_norm": 0.1698072850704193, | |
| "learning_rate": 1.7274575140626318e-05, | |
| "loss": 0.1691, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.03808487486398259, | |
| "grad_norm": 0.16231591999530792, | |
| "learning_rate": 1.5432914190872757e-05, | |
| "loss": 0.1603, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.03917301414581066, | |
| "grad_norm": 0.16352759301662445, | |
| "learning_rate": 1.3650237506511331e-05, | |
| "loss": 0.1365, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.04026115342763874, | |
| "grad_norm": 0.2394663244485855, | |
| "learning_rate": 1.1937535882101281e-05, | |
| "loss": 0.2083, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.041349292709466814, | |
| "grad_norm": 0.2228260487318039, | |
| "learning_rate": 1.0305368692688174e-05, | |
| "loss": 0.1766, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.042437431991294884, | |
| "grad_norm": 0.16056722402572632, | |
| "learning_rate": 8.763798791745411e-06, | |
| "loss": 0.1458, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.042437431991294884, | |
| "eval_loss": 0.18522028625011444, | |
| "eval_runtime": 20.7477, | |
| "eval_samples_per_second": 18.653, | |
| "eval_steps_per_second": 9.35, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.04352557127312296, | |
| "grad_norm": 0.20927105844020844, | |
| "learning_rate": 7.3223304703363135e-06, | |
| "loss": 0.1808, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.04461371055495103, | |
| "grad_norm": 0.19594742357730865, | |
| "learning_rate": 5.989850859999227e-06, | |
| "loss": 0.1499, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.04570184983677911, | |
| "grad_norm": 0.3185279667377472, | |
| "learning_rate": 4.7745751406263165e-06, | |
| "loss": 0.2901, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.046789989118607184, | |
| "grad_norm": 0.19601090252399445, | |
| "learning_rate": 3.6839958911476957e-06, | |
| "loss": 0.1651, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.04787812840043525, | |
| "grad_norm": 0.3092597723007202, | |
| "learning_rate": 2.7248368952908053e-06, | |
| "loss": 0.1719, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.04896626768226333, | |
| "grad_norm": 0.15384280681610107, | |
| "learning_rate": 1.9030116872178316e-06, | |
| "loss": 0.13, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.05005440696409141, | |
| "grad_norm": 0.22593477368354797, | |
| "learning_rate": 1.2235870926211619e-06, | |
| "loss": 0.1633, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.051142546245919476, | |
| "grad_norm": 0.1992155909538269, | |
| "learning_rate": 6.907519900580861e-07, | |
| "loss": 0.1572, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.05223068552774755, | |
| "grad_norm": 0.21265120804309845, | |
| "learning_rate": 3.077914851215585e-07, | |
| "loss": 0.1135, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.05331882480957562, | |
| "grad_norm": 0.16999712586402893, | |
| "learning_rate": 7.706665667180091e-08, | |
| "loss": 0.144, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0544069640914037, | |
| "grad_norm": 0.2375941425561905, | |
| "learning_rate": 0.0, | |
| "loss": 0.1357, | |
| "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": 7687956234240000.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |