| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.5309734513274336, | |
| "eval_steps": 50, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11799410029498525, | |
| "grad_norm": 2.4879140853881836, | |
| "learning_rate": 0.0003, | |
| "loss": 1.8567, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2359882005899705, | |
| "grad_norm": 0.8311281204223633, | |
| "learning_rate": 0.0002963855421686747, | |
| "loss": 1.3245, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.35398230088495575, | |
| "grad_norm": 0.6665307879447937, | |
| "learning_rate": 0.0002927710843373494, | |
| "loss": 1.1834, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.471976401179941, | |
| "grad_norm": 0.5632417798042297, | |
| "learning_rate": 0.0002891566265060241, | |
| "loss": 1.0915, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5899705014749262, | |
| "grad_norm": 0.4381982684135437, | |
| "learning_rate": 0.00028554216867469873, | |
| "loss": 1.0438, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.5899705014749262, | |
| "eval_loss": 1.0292344093322754, | |
| "eval_runtime": 20.4608, | |
| "eval_samples_per_second": 8.015, | |
| "eval_steps_per_second": 0.684, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.7079646017699115, | |
| "grad_norm": 0.43256932497024536, | |
| "learning_rate": 0.0002819277108433735, | |
| "loss": 1.0156, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.8259587020648967, | |
| "grad_norm": 0.3582414388656616, | |
| "learning_rate": 0.0002783132530120482, | |
| "loss": 0.9793, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.943952802359882, | |
| "grad_norm": 0.3671708405017853, | |
| "learning_rate": 0.00027469879518072284, | |
| "loss": 0.9644, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.0589970501474926, | |
| "grad_norm": 0.34048739075660706, | |
| "learning_rate": 0.0002710843373493976, | |
| "loss": 0.9173, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.176991150442478, | |
| "grad_norm": 0.29820743203163147, | |
| "learning_rate": 0.00026746987951807225, | |
| "loss": 0.8913, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.176991150442478, | |
| "eval_loss": 0.9288875460624695, | |
| "eval_runtime": 20.3929, | |
| "eval_samples_per_second": 8.042, | |
| "eval_steps_per_second": 0.687, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.294985250737463, | |
| "grad_norm": 0.31813672184944153, | |
| "learning_rate": 0.00026385542168674695, | |
| "loss": 0.8801, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.4129793510324484, | |
| "grad_norm": 0.2814970016479492, | |
| "learning_rate": 0.00026024096385542165, | |
| "loss": 0.875, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.5309734513274336, | |
| "grad_norm": 0.2815490663051605, | |
| "learning_rate": 0.00025662650602409636, | |
| "loss": 0.8642, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.648967551622419, | |
| "grad_norm": 0.2590789198875427, | |
| "learning_rate": 0.00025301204819277106, | |
| "loss": 0.8501, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.7669616519174043, | |
| "grad_norm": 0.25773778557777405, | |
| "learning_rate": 0.00024939759036144576, | |
| "loss": 0.8487, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.7669616519174043, | |
| "eval_loss": 0.8774459362030029, | |
| "eval_runtime": 20.4395, | |
| "eval_samples_per_second": 8.024, | |
| "eval_steps_per_second": 0.685, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.8849557522123894, | |
| "grad_norm": 0.2629424035549164, | |
| "learning_rate": 0.00024578313253012046, | |
| "loss": 0.8427, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.2865707278251648, | |
| "learning_rate": 0.00024216867469879517, | |
| "loss": 0.8258, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.117994100294985, | |
| "grad_norm": 0.27073633670806885, | |
| "learning_rate": 0.00023855421686746987, | |
| "loss": 0.7835, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.2359882005899703, | |
| "grad_norm": 0.2692751884460449, | |
| "learning_rate": 0.00023493975903614455, | |
| "loss": 0.7804, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.353982300884956, | |
| "grad_norm": 0.2548248767852783, | |
| "learning_rate": 0.00023132530120481928, | |
| "loss": 0.7756, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.353982300884956, | |
| "eval_loss": 0.8462932109832764, | |
| "eval_runtime": 20.4134, | |
| "eval_samples_per_second": 8.034, | |
| "eval_steps_per_second": 0.686, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.471976401179941, | |
| "grad_norm": 0.21938325464725494, | |
| "learning_rate": 0.00022771084337349395, | |
| "loss": 0.7695, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.589970501474926, | |
| "grad_norm": 0.23754891753196716, | |
| "learning_rate": 0.00022409638554216866, | |
| "loss": 0.7771, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.7079646017699117, | |
| "grad_norm": 0.2382248044013977, | |
| "learning_rate": 0.00022048192771084336, | |
| "loss": 0.7672, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.825958702064897, | |
| "grad_norm": 0.2536454200744629, | |
| "learning_rate": 0.00021686746987951806, | |
| "loss": 0.7645, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.943952802359882, | |
| "grad_norm": 0.21255475282669067, | |
| "learning_rate": 0.00021325301204819274, | |
| "loss": 0.7667, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.943952802359882, | |
| "eval_loss": 0.8220995664596558, | |
| "eval_runtime": 20.3922, | |
| "eval_samples_per_second": 8.042, | |
| "eval_steps_per_second": 0.687, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 3.0589970501474926, | |
| "grad_norm": 0.2579372227191925, | |
| "learning_rate": 0.00020963855421686747, | |
| "loss": 0.7418, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 3.1769911504424777, | |
| "grad_norm": 0.2223750501871109, | |
| "learning_rate": 0.00020602409638554214, | |
| "loss": 0.7198, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 3.2949852507374633, | |
| "grad_norm": 0.26154494285583496, | |
| "learning_rate": 0.00020240963855421685, | |
| "loss": 0.7188, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 3.4129793510324484, | |
| "grad_norm": 0.2428540140390396, | |
| "learning_rate": 0.00019879518072289155, | |
| "loss": 0.7158, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.5309734513274336, | |
| "grad_norm": 0.2056645303964615, | |
| "learning_rate": 0.00019518072289156625, | |
| "loss": 0.7103, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.5309734513274336, | |
| "eval_loss": 0.8080394864082336, | |
| "eval_runtime": 20.4084, | |
| "eval_samples_per_second": 8.036, | |
| "eval_steps_per_second": 0.686, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 840, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.433776809382511e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |