| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 14.980891719745223, | |
| "eval_steps": 1, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.752838134765625, | |
| "learning_rate": 0.00013103448275862068, | |
| "loss": 1.9056, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "eval_loss": 1.2314808368682861, | |
| "eval_runtime": 41.0888, | |
| "eval_samples_per_second": 3.407, | |
| "eval_steps_per_second": 3.407, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 0.3157634139060974, | |
| "learning_rate": 0.000197828447339848, | |
| "loss": 0.7969, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_loss": 0.6621134877204895, | |
| "eval_runtime": 41.1024, | |
| "eval_samples_per_second": 3.406, | |
| "eval_steps_per_second": 3.406, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 0.46234801411628723, | |
| "learning_rate": 0.0001937024972855592, | |
| "loss": 0.564, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "eval_loss": 0.5718151330947876, | |
| "eval_runtime": 41.1043, | |
| "eval_samples_per_second": 3.406, | |
| "eval_steps_per_second": 3.406, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "grad_norm": 0.5510583519935608, | |
| "learning_rate": 0.00018935939196525515, | |
| "loss": 0.3962, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 3.97, | |
| "eval_loss": 0.5229483246803284, | |
| "eval_runtime": 41.1671, | |
| "eval_samples_per_second": 3.401, | |
| "eval_steps_per_second": 3.401, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "grad_norm": 4.286730766296387, | |
| "learning_rate": 0.00018501628664495113, | |
| "loss": 0.3022, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_loss": 0.5116992592811584, | |
| "eval_runtime": 41.1636, | |
| "eval_samples_per_second": 3.401, | |
| "eval_steps_per_second": 3.401, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "grad_norm": 0.6076149940490723, | |
| "learning_rate": 0.00018089033659066234, | |
| "loss": 0.2504, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "eval_loss": 0.5122354030609131, | |
| "eval_runtime": 41.1631, | |
| "eval_samples_per_second": 3.401, | |
| "eval_steps_per_second": 3.401, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "grad_norm": 0.5720144510269165, | |
| "learning_rate": 0.0001765472312703583, | |
| "loss": 0.1934, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_loss": 0.5105646252632141, | |
| "eval_runtime": 41.1792, | |
| "eval_samples_per_second": 3.4, | |
| "eval_steps_per_second": 3.4, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.6266606450080872, | |
| "learning_rate": 0.0001722041259500543, | |
| "loss": 0.1632, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.523033082485199, | |
| "eval_runtime": 41.2092, | |
| "eval_samples_per_second": 3.397, | |
| "eval_steps_per_second": 3.397, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "grad_norm": 0.6278405785560608, | |
| "learning_rate": 0.00016807817589576548, | |
| "loss": 0.147, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 8.97, | |
| "eval_loss": 0.5415217876434326, | |
| "eval_runtime": 41.2395, | |
| "eval_samples_per_second": 3.395, | |
| "eval_steps_per_second": 3.395, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "grad_norm": 0.5813736319541931, | |
| "learning_rate": 0.00016373507057546147, | |
| "loss": 0.1238, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_loss": 0.548870861530304, | |
| "eval_runtime": 41.253, | |
| "eval_samples_per_second": 3.394, | |
| "eval_steps_per_second": 3.394, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "grad_norm": 0.48950037360191345, | |
| "learning_rate": 0.00015960912052117264, | |
| "loss": 0.1162, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 10.96, | |
| "eval_loss": 0.5625630021095276, | |
| "eval_runtime": 41.2366, | |
| "eval_samples_per_second": 3.395, | |
| "eval_steps_per_second": 3.395, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "grad_norm": 0.45576944947242737, | |
| "learning_rate": 0.00015526601520086863, | |
| "loss": 0.1003, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 11.97, | |
| "eval_loss": 0.5758817195892334, | |
| "eval_runtime": 41.2495, | |
| "eval_samples_per_second": 3.394, | |
| "eval_steps_per_second": 3.394, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "grad_norm": 0.4926683008670807, | |
| "learning_rate": 0.0001509229098805646, | |
| "loss": 0.0945, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_loss": 0.5838292837142944, | |
| "eval_runtime": 41.2818, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 3.391, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "grad_norm": 0.4472861588001251, | |
| "learning_rate": 0.0001467969598262758, | |
| "loss": 0.0945, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 13.96, | |
| "eval_loss": 0.5958499312400818, | |
| "eval_runtime": 41.2896, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 3.391, | |
| "step": 274 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "grad_norm": 0.3718027174472809, | |
| "learning_rate": 0.00014245385450597177, | |
| "loss": 0.0858, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_loss": 0.5897623300552368, | |
| "eval_runtime": 41.289, | |
| "eval_samples_per_second": 3.391, | |
| "eval_steps_per_second": 3.391, | |
| "step": 294 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 950, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 1, | |
| "total_flos": 9.871664050142208e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |