| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 7.0, |
| "eval_steps": 500, |
| "global_step": 77, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.46875, |
| "grad_norm": 3.5517272105283895, |
| "learning_rate": 2e-05, |
| "loss": 0.4691, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.1389603316783905, |
| "step": 5, |
| "valid_targets_mean": 8612.9, |
| "valid_targets_min": 1973 |
| }, |
| { |
| "epoch": 0.9375, |
| "grad_norm": 1.6035954300006818, |
| "learning_rate": 3.9979273472934556e-05, |
| "loss": 0.3617, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.1101037785410881, |
| "step": 10, |
| "valid_targets_mean": 10068.0, |
| "valid_targets_min": 6315 |
| }, |
| { |
| "epoch": 1.375, |
| "grad_norm": 0.4264672102775543, |
| "learning_rate": 3.925834574695599e-05, |
| "loss": 0.3138, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.09861762076616287, |
| "step": 15, |
| "valid_targets_mean": 9279.8, |
| "valid_targets_min": 2555 |
| }, |
| { |
| "epoch": 1.84375, |
| "grad_norm": 0.3215489540810774, |
| "learning_rate": 3.754365204805189e-05, |
| "loss": 0.266, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.0835193321108818, |
| "step": 20, |
| "valid_targets_mean": 8479.3, |
| "valid_targets_min": 2433 |
| }, |
| { |
| "epoch": 2.28125, |
| "grad_norm": 0.2588293119372081, |
| "learning_rate": 3.4923673716953717e-05, |
| "loss": 0.2393, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.07589198648929596, |
| "step": 25, |
| "valid_targets_mean": 8381.5, |
| "valid_targets_min": 3027 |
| }, |
| { |
| "epoch": 2.75, |
| "grad_norm": 0.23592440501636647, |
| "learning_rate": 3.153360644229735e-05, |
| "loss": 0.2217, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.07492702454328537, |
| "step": 30, |
| "valid_targets_mean": 8741.6, |
| "valid_targets_min": 2273 |
| }, |
| { |
| "epoch": 3.1875, |
| "grad_norm": 0.20000001943996298, |
| "learning_rate": 2.7548383915052287e-05, |
| "loss": 0.2148, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.07082916051149368, |
| "step": 35, |
| "valid_targets_mean": 7941.8, |
| "valid_targets_min": 2787 |
| }, |
| { |
| "epoch": 3.65625, |
| "grad_norm": 0.17026247891957802, |
| "learning_rate": 2.317365092912503e-05, |
| "loss": 0.2016, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.07093578577041626, |
| "step": 40, |
| "valid_targets_mean": 9171.9, |
| "valid_targets_min": 1973 |
| }, |
| { |
| "epoch": 4.09375, |
| "grad_norm": 0.16370983829999688, |
| "learning_rate": 1.8635151732706586e-05, |
| "loss": 0.1938, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.06410802900791168, |
| "step": 45, |
| "valid_targets_mean": 9452.5, |
| "valid_targets_min": 5050 |
| }, |
| { |
| "epoch": 4.5625, |
| "grad_norm": 0.15161029449631325, |
| "learning_rate": 1.4167081210903501e-05, |
| "loss": 0.1958, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.07220222800970078, |
| "step": 50, |
| "valid_targets_mean": 9556.2, |
| "valid_targets_min": 3856 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.19420589194134402, |
| "learning_rate": 1.0000000000000006e-05, |
| "loss": 0.1849, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.08684807270765305, |
| "step": 55, |
| "valid_targets_mean": 8111.1, |
| "valid_targets_min": 2073 |
| }, |
| { |
| "epoch": 5.46875, |
| "grad_norm": 0.14890475467039743, |
| "learning_rate": 6.348937135626922e-06, |
| "loss": 0.186, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.06543072313070297, |
| "step": 60, |
| "valid_targets_mean": 9081.7, |
| "valid_targets_min": 2927 |
| }, |
| { |
| "epoch": 5.9375, |
| "grad_norm": 0.15377574900801283, |
| "learning_rate": 3.402294160019499e-06, |
| "loss": 0.1845, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.060948729515075684, |
| "step": 65, |
| "valid_targets_mean": 8957.2, |
| "valid_targets_min": 4178 |
| }, |
| { |
| "epoch": 6.375, |
| "grad_norm": 0.1463889873558677, |
| "learning_rate": 1.3121232567865793e-06, |
| "loss": 0.1832, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.06596855819225311, |
| "step": 70, |
| "valid_targets_mean": 9267.8, |
| "valid_targets_min": 2315 |
| }, |
| { |
| "epoch": 6.84375, |
| "grad_norm": 0.15364253952679346, |
| "learning_rate": 1.862810792733849e-07, |
| "loss": 0.1819, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.06382087618112564, |
| "step": 75, |
| "valid_targets_mean": 9101.6, |
| "valid_targets_min": 1782 |
| }, |
| { |
| "epoch": 7.0, |
| "loss_nan_ranks": 0, |
| "loss_rank_avg": 0.09879625588655472, |
| "step": 77, |
| "total_flos": 5.4876432713016934e+17, |
| "train_loss": 0.23884032221583576, |
| "train_runtime": 2368.0537, |
| "train_samples_per_second": 2.956, |
| "train_steps_per_second": 0.033, |
| "valid_targets_mean": 9211.3, |
| "valid_targets_min": 4302 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 77, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 7, |
| "save_steps": 300, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 5.4876432713016934e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
