| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.9391304347826086, | |
| "eval_steps": 4, | |
| "global_step": 28, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06956521739130435, | |
| "grad_norm": 0.24870926141738892, | |
| "learning_rate": 2e-05, | |
| "loss": 0.9615, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.06956521739130435, | |
| "eval_loss": 0.9210608005523682, | |
| "eval_runtime": 74.7303, | |
| "eval_samples_per_second": 2.088, | |
| "eval_steps_per_second": 1.044, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1391304347826087, | |
| "grad_norm": 0.2108100801706314, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7833, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.20869565217391303, | |
| "grad_norm": 0.1993255764245987, | |
| "learning_rate": 6e-05, | |
| "loss": 0.7999, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "grad_norm": 0.28567054867744446, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8231, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2782608695652174, | |
| "eval_loss": 0.9139525294303894, | |
| "eval_runtime": 75.4634, | |
| "eval_samples_per_second": 2.067, | |
| "eval_steps_per_second": 1.034, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.2238488644361496, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6855, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.41739130434782606, | |
| "grad_norm": 0.248153418302536, | |
| "learning_rate": 0.00012, | |
| "loss": 0.9259, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.48695652173913045, | |
| "grad_norm": 0.252511590719223, | |
| "learning_rate": 0.00014, | |
| "loss": 0.7668, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "grad_norm": 0.25402677059173584, | |
| "learning_rate": 0.00016, | |
| "loss": 0.779, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5565217391304348, | |
| "eval_loss": 0.8034127354621887, | |
| "eval_runtime": 75.4244, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 1.034, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.6260869565217392, | |
| "grad_norm": 0.1748448610305786, | |
| "learning_rate": 0.00018, | |
| "loss": 0.5345, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.18634748458862305, | |
| "learning_rate": 0.0002, | |
| "loss": 0.5699, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.7652173913043478, | |
| "grad_norm": 0.21753615140914917, | |
| "learning_rate": 0.00019848077530122083, | |
| "loss": 0.5804, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "grad_norm": 0.23333732783794403, | |
| "learning_rate": 0.00019396926207859084, | |
| "loss": 0.545, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.8347826086956521, | |
| "eval_loss": 0.648797869682312, | |
| "eval_runtime": 75.308, | |
| "eval_samples_per_second": 2.071, | |
| "eval_steps_per_second": 1.036, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.9043478260869565, | |
| "grad_norm": 0.20313376188278198, | |
| "learning_rate": 0.00018660254037844388, | |
| "loss": 0.4392, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.9739130434782609, | |
| "grad_norm": 0.20026056468486786, | |
| "learning_rate": 0.0001766044443118978, | |
| "loss": 0.38, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.0347826086956522, | |
| "grad_norm": 0.153587207198143, | |
| "learning_rate": 0.00016427876096865394, | |
| "loss": 0.3177, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.1043478260869566, | |
| "grad_norm": 0.16564497351646423, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.3712, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.1043478260869566, | |
| "eval_loss": 0.5825337767601013, | |
| "eval_runtime": 75.4662, | |
| "eval_samples_per_second": 2.067, | |
| "eval_steps_per_second": 1.034, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.1739130434782608, | |
| "grad_norm": 0.1881798803806305, | |
| "learning_rate": 0.00013420201433256689, | |
| "loss": 0.3357, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.2434782608695651, | |
| "grad_norm": 0.21584919095039368, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 0.3875, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.3130434782608695, | |
| "grad_norm": 0.1652679145336151, | |
| "learning_rate": 0.0001, | |
| "loss": 0.287, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.382608695652174, | |
| "grad_norm": 0.1664346307516098, | |
| "learning_rate": 8.263518223330697e-05, | |
| "loss": 0.336, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.382608695652174, | |
| "eval_loss": 0.5491526126861572, | |
| "eval_runtime": 75.3653, | |
| "eval_samples_per_second": 2.07, | |
| "eval_steps_per_second": 1.035, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.4521739130434783, | |
| "grad_norm": 0.15922914445400238, | |
| "learning_rate": 6.579798566743314e-05, | |
| "loss": 0.2959, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.5217391304347827, | |
| "grad_norm": 0.14202940464019775, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.2983, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.591304347826087, | |
| "grad_norm": 0.1265108287334442, | |
| "learning_rate": 3.5721239031346066e-05, | |
| "loss": 0.247, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.6608695652173913, | |
| "grad_norm": 0.15941433608531952, | |
| "learning_rate": 2.339555568810221e-05, | |
| "loss": 0.3231, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.6608695652173913, | |
| "eval_loss": 0.537909984588623, | |
| "eval_runtime": 75.2911, | |
| "eval_samples_per_second": 2.072, | |
| "eval_steps_per_second": 1.036, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.7304347826086957, | |
| "grad_norm": 0.1614454984664917, | |
| "learning_rate": 1.339745962155613e-05, | |
| "loss": 0.311, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.13931332528591156, | |
| "learning_rate": 6.030737921409169e-06, | |
| "loss": 0.2731, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.8695652173913042, | |
| "grad_norm": 0.15171031653881073, | |
| "learning_rate": 1.5192246987791981e-06, | |
| "loss": 0.2874, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.9391304347826086, | |
| "grad_norm": 0.14994187653064728, | |
| "learning_rate": 0.0, | |
| "loss": 0.2729, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.9391304347826086, | |
| "eval_loss": 0.5353411436080933, | |
| "eval_runtime": 75.3284, | |
| "eval_samples_per_second": 2.071, | |
| "eval_steps_per_second": 1.035, | |
| "step": 28 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 28, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 14, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.149946891848909e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |