| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0785391714117416, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0007853917141174161, | |
| "eval_loss": 8.724822998046875, | |
| "eval_runtime": 233.5003, | |
| "eval_samples_per_second": 19.392, | |
| "eval_steps_per_second": 2.424, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0039269585705870805, | |
| "grad_norm": 6.884418964385986, | |
| "learning_rate": 3.75e-05, | |
| "loss": 8.547, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.007068525427056745, | |
| "eval_loss": 6.619219779968262, | |
| "eval_runtime": 233.7681, | |
| "eval_samples_per_second": 19.37, | |
| "eval_steps_per_second": 2.421, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.007853917141174161, | |
| "grad_norm": 9.305004119873047, | |
| "learning_rate": 7.5e-05, | |
| "loss": 7.0279, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.011780875711761242, | |
| "grad_norm": 8.224003791809082, | |
| "learning_rate": 0.0001125, | |
| "loss": 5.2814, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.01413705085411349, | |
| "eval_loss": 4.465071678161621, | |
| "eval_runtime": 233.7503, | |
| "eval_samples_per_second": 19.371, | |
| "eval_steps_per_second": 2.421, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.015707834282348322, | |
| "grad_norm": 8.007319450378418, | |
| "learning_rate": 0.00015, | |
| "loss": 4.5609, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0196347928529354, | |
| "grad_norm": 8.696537017822266, | |
| "learning_rate": 0.00014855889603024227, | |
| "loss": 4.2927, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.021205576281170233, | |
| "eval_loss": 4.00045919418335, | |
| "eval_runtime": 233.7374, | |
| "eval_samples_per_second": 19.372, | |
| "eval_steps_per_second": 2.422, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.023561751423522483, | |
| "grad_norm": 6.938048839569092, | |
| "learning_rate": 0.0001442909649383465, | |
| "loss": 4.0025, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.027488709994109562, | |
| "grad_norm": 7.009843826293945, | |
| "learning_rate": 0.0001373602209226909, | |
| "loss": 3.8531, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.02827410170822698, | |
| "eval_loss": 3.82582426071167, | |
| "eval_runtime": 233.7172, | |
| "eval_samples_per_second": 19.374, | |
| "eval_steps_per_second": 2.422, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.031415668564696644, | |
| "grad_norm": 6.201370716094971, | |
| "learning_rate": 0.00012803300858899104, | |
| "loss": 3.719, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03534262713528372, | |
| "grad_norm": 5.8150105476379395, | |
| "learning_rate": 0.00011666776747647015, | |
| "loss": 3.6351, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.03534262713528372, | |
| "eval_loss": 3.6261699199676514, | |
| "eval_runtime": 233.6588, | |
| "eval_samples_per_second": 19.379, | |
| "eval_steps_per_second": 2.422, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0392695857058708, | |
| "grad_norm": 4.917875289916992, | |
| "learning_rate": 0.00010370125742738173, | |
| "loss": 3.6745, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.042411152562340466, | |
| "eval_loss": 3.5405988693237305, | |
| "eval_runtime": 233.7238, | |
| "eval_samples_per_second": 19.373, | |
| "eval_steps_per_second": 2.422, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.04319654427645788, | |
| "grad_norm": 4.494648456573486, | |
| "learning_rate": 8.963177415120962e-05, | |
| "loss": 3.5363, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.04712350284704497, | |
| "grad_norm": 6.017573833465576, | |
| "learning_rate": 7.5e-05, | |
| "loss": 3.2906, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.04947967798939721, | |
| "eval_loss": 3.450070381164551, | |
| "eval_runtime": 233.7549, | |
| "eval_samples_per_second": 19.371, | |
| "eval_steps_per_second": 2.421, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.051050461417632045, | |
| "grad_norm": 4.84181022644043, | |
| "learning_rate": 6.036822584879038e-05, | |
| "loss": 3.3757, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.054977419988219124, | |
| "grad_norm": 6.739719390869141, | |
| "learning_rate": 4.6298742572618266e-05, | |
| "loss": 3.4344, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.05654820341645396, | |
| "eval_loss": 3.3793962001800537, | |
| "eval_runtime": 233.8278, | |
| "eval_samples_per_second": 19.365, | |
| "eval_steps_per_second": 2.421, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0589043785588062, | |
| "grad_norm": 4.579084873199463, | |
| "learning_rate": 3.333223252352985e-05, | |
| "loss": 3.3768, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.06283133712939329, | |
| "grad_norm": 5.169261932373047, | |
| "learning_rate": 2.1966991411008938e-05, | |
| "loss": 3.4602, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.0636167288435107, | |
| "eval_loss": 3.332181453704834, | |
| "eval_runtime": 233.7069, | |
| "eval_samples_per_second": 19.375, | |
| "eval_steps_per_second": 2.422, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.06675829569998036, | |
| "grad_norm": 5.313036918640137, | |
| "learning_rate": 1.2639779077309098e-05, | |
| "loss": 3.3791, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.07068525427056745, | |
| "grad_norm": 4.830722332000732, | |
| "learning_rate": 5.709035061653494e-06, | |
| "loss": 3.0172, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07068525427056745, | |
| "eval_loss": 3.3156657218933105, | |
| "eval_runtime": 233.6444, | |
| "eval_samples_per_second": 19.38, | |
| "eval_steps_per_second": 2.422, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07461221284115453, | |
| "grad_norm": 5.533350944519043, | |
| "learning_rate": 1.4411039697577175e-06, | |
| "loss": 3.4561, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.07775377969762419, | |
| "eval_loss": 3.3061177730560303, | |
| "eval_runtime": 233.6519, | |
| "eval_samples_per_second": 19.379, | |
| "eval_steps_per_second": 2.422, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.0785391714117416, | |
| "grad_norm": 4.896818161010742, | |
| "learning_rate": 0.0, | |
| "loss": 2.9994, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.508513578745856e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |