| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.004218697266284171, | |
| "eval_steps": 9, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00016874789065136686, | |
| "grad_norm": 0.6673353910446167, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1828, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00016874789065136686, | |
| "eval_loss": 1.133446216583252, | |
| "eval_runtime": 670.9071, | |
| "eval_samples_per_second": 7.438, | |
| "eval_steps_per_second": 0.93, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003374957813027337, | |
| "grad_norm": 0.6421517133712769, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1352, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0005062436719541006, | |
| "grad_norm": 0.8541758060455322, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9903, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0006749915626054674, | |
| "grad_norm": 0.4865100383758545, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1104, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0008437394532568343, | |
| "grad_norm": 0.5594805479049683, | |
| "learning_rate": 5e-05, | |
| "loss": 0.9953, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0010124873439082012, | |
| "grad_norm": 0.5936105251312256, | |
| "learning_rate": 6e-05, | |
| "loss": 1.019, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.001181235234559568, | |
| "grad_norm": 0.7241169214248657, | |
| "learning_rate": 7e-05, | |
| "loss": 1.0544, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0013499831252109348, | |
| "grad_norm": 0.8012577295303345, | |
| "learning_rate": 8e-05, | |
| "loss": 1.0349, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0015187310158623017, | |
| "grad_norm": 0.6014957427978516, | |
| "learning_rate": 9e-05, | |
| "loss": 0.8921, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0015187310158623017, | |
| "eval_loss": 0.9576449990272522, | |
| "eval_runtime": 672.5267, | |
| "eval_samples_per_second": 7.42, | |
| "eval_steps_per_second": 0.928, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0016874789065136687, | |
| "grad_norm": 0.3161397874355316, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0526, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0018562267971650355, | |
| "grad_norm": 0.3254038393497467, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.9626, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0020249746878164025, | |
| "grad_norm": 0.29402822256088257, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8483, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0021937225784677693, | |
| "grad_norm": 0.3435806930065155, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.9679, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.002362470469119136, | |
| "grad_norm": 0.35293349623680115, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.9465, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.002531218359770503, | |
| "grad_norm": 0.3628636598587036, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.832, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0026999662504218697, | |
| "grad_norm": 0.354307621717453, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7619, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0028687141410732365, | |
| "grad_norm": 0.40063461661338806, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.8526, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0030374620317246033, | |
| "grad_norm": 0.3604790270328522, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.8343, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0030374620317246033, | |
| "eval_loss": 0.8436176180839539, | |
| "eval_runtime": 672.8891, | |
| "eval_samples_per_second": 7.416, | |
| "eval_steps_per_second": 0.927, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00320620992237597, | |
| "grad_norm": 0.32435283064842224, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.8905, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0033749578130273373, | |
| "grad_norm": 0.29493603110313416, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.8299, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.003543705703678704, | |
| "grad_norm": 0.28454816341400146, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.8245, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.003712453594330071, | |
| "grad_norm": 0.30943214893341064, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.8342, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0038812014849814377, | |
| "grad_norm": 0.2970975637435913, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.854, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.004049949375632805, | |
| "grad_norm": 0.32714584469795227, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8385, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.004218697266284171, | |
| "grad_norm": 0.30870357155799866, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.715, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.049391762341888e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |