| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1266624445851805, | |
| "eval_steps": 3, | |
| "global_step": 25, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00506649778340722, | |
| "grad_norm": 4.028384208679199, | |
| "learning_rate": 2e-05, | |
| "loss": 3.5544, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00506649778340722, | |
| "eval_loss": 3.5219967365264893, | |
| "eval_runtime": 17.0746, | |
| "eval_samples_per_second": 9.781, | |
| "eval_steps_per_second": 4.92, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.01013299556681444, | |
| "grad_norm": 4.173046588897705, | |
| "learning_rate": 4e-05, | |
| "loss": 3.9016, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.015199493350221659, | |
| "grad_norm": 3.8153645992279053, | |
| "learning_rate": 6e-05, | |
| "loss": 3.3707, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.015199493350221659, | |
| "eval_loss": 3.4256696701049805, | |
| "eval_runtime": 17.0755, | |
| "eval_samples_per_second": 9.78, | |
| "eval_steps_per_second": 4.919, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.02026599113362888, | |
| "grad_norm": 3.7728078365325928, | |
| "learning_rate": 8e-05, | |
| "loss": 3.2427, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0253324889170361, | |
| "grad_norm": 4.0927228927612305, | |
| "learning_rate": 0.0001, | |
| "loss": 2.8313, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.030398986700443317, | |
| "grad_norm": 4.602127552032471, | |
| "learning_rate": 0.00012, | |
| "loss": 3.0358, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.030398986700443317, | |
| "eval_loss": 2.225193500518799, | |
| "eval_runtime": 17.159, | |
| "eval_samples_per_second": 9.733, | |
| "eval_steps_per_second": 4.895, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.03546548448385054, | |
| "grad_norm": 4.011180877685547, | |
| "learning_rate": 0.00014, | |
| "loss": 1.9538, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.04053198226725776, | |
| "grad_norm": 3.0159177780151367, | |
| "learning_rate": 0.00016, | |
| "loss": 1.4574, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.045598480050664976, | |
| "grad_norm": 3.2096781730651855, | |
| "learning_rate": 0.00018, | |
| "loss": 1.4471, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.045598480050664976, | |
| "eval_loss": 1.0822527408599854, | |
| "eval_runtime": 17.0804, | |
| "eval_samples_per_second": 9.777, | |
| "eval_steps_per_second": 4.918, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0506649778340722, | |
| "grad_norm": 3.9092917442321777, | |
| "learning_rate": 0.0002, | |
| "loss": 1.0191, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.05573147561747942, | |
| "grad_norm": 3.5135035514831543, | |
| "learning_rate": 0.00019781476007338058, | |
| "loss": 0.4289, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.060797973400886635, | |
| "grad_norm": 3.6494600772857666, | |
| "learning_rate": 0.0001913545457642601, | |
| "loss": 0.3012, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.060797973400886635, | |
| "eval_loss": 0.1743772327899933, | |
| "eval_runtime": 17.1197, | |
| "eval_samples_per_second": 9.755, | |
| "eval_steps_per_second": 4.907, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.06586447118429385, | |
| "grad_norm": 1.6839953660964966, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.1816, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.07093096896770108, | |
| "grad_norm": 1.779843807220459, | |
| "learning_rate": 0.00016691306063588583, | |
| "loss": 0.1379, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0759974667511083, | |
| "grad_norm": 3.1982781887054443, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2963, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0759974667511083, | |
| "eval_loss": 0.11545062810182571, | |
| "eval_runtime": 17.145, | |
| "eval_samples_per_second": 9.74, | |
| "eval_steps_per_second": 4.899, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.08106396453451552, | |
| "grad_norm": 0.8417478203773499, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.0254, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.08613046231792273, | |
| "grad_norm": 1.9363985061645508, | |
| "learning_rate": 0.00011045284632676536, | |
| "loss": 0.1866, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.09119696010132995, | |
| "grad_norm": 1.8391631841659546, | |
| "learning_rate": 8.954715367323468e-05, | |
| "loss": 0.1333, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09119696010132995, | |
| "eval_loss": 0.12381552159786224, | |
| "eval_runtime": 17.1973, | |
| "eval_samples_per_second": 9.711, | |
| "eval_steps_per_second": 4.884, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.09626345788473717, | |
| "grad_norm": 1.8057817220687866, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.1361, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.1013299556681444, | |
| "grad_norm": 1.080073595046997, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0566, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.10639645345155162, | |
| "grad_norm": 0.8825861811637878, | |
| "learning_rate": 3.308693936411421e-05, | |
| "loss": 0.056, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.10639645345155162, | |
| "eval_loss": 0.13038882613182068, | |
| "eval_runtime": 17.2129, | |
| "eval_samples_per_second": 9.702, | |
| "eval_steps_per_second": 4.88, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.11146295123495883, | |
| "grad_norm": 0.5869631171226501, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.0261, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.11652944901836605, | |
| "grad_norm": 0.7820150852203369, | |
| "learning_rate": 8.645454235739903e-06, | |
| "loss": 0.0398, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.12159594680177327, | |
| "grad_norm": 0.20869500935077667, | |
| "learning_rate": 2.1852399266194314e-06, | |
| "loss": 0.0073, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.12159594680177327, | |
| "eval_loss": 0.129783034324646, | |
| "eval_runtime": 17.2227, | |
| "eval_samples_per_second": 9.696, | |
| "eval_steps_per_second": 4.877, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.1266624445851805, | |
| "grad_norm": 0.2913738489151001, | |
| "learning_rate": 0.0, | |
| "loss": 0.01, | |
| "step": 25 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 25, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.3054705844224e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |