| { | |
| "best_metric": 0.7917657494544983, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_60-samples_config-2_auto/checkpoint-14", | |
| "epoch": 11.826086956521738, | |
| "eval_steps": 500, | |
| "global_step": 34, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 2.5684735774993896, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1404, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.6617414951324463, | |
| "learning_rate": 2e-05, | |
| "loss": 1.0805, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.081933617591858, | |
| "eval_runtime": 5.1346, | |
| "eval_samples_per_second": 2.337, | |
| "eval_steps_per_second": 2.337, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 2.1907503604888916, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0214, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 0.9261219501495361, | |
| "eval_runtime": 5.1308, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 1.6325212717056274, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9786, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.114859938621521, | |
| "learning_rate": 8e-05, | |
| "loss": 0.7668, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 0.8321869373321533, | |
| "eval_runtime": 5.134, | |
| "eval_samples_per_second": 2.337, | |
| "eval_steps_per_second": 2.337, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.306771993637085, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7312, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 0.7920494079589844, | |
| "eval_runtime": 5.1311, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 1.3007943630218506, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6821, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.8811099529266357, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.5041, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.7917657494544983, | |
| "eval_runtime": 5.1316, | |
| "eval_samples_per_second": 2.338, | |
| "eval_steps_per_second": 2.338, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 1.021318793296814, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.4287, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8764511942863464, | |
| "eval_runtime": 5.1363, | |
| "eval_samples_per_second": 2.336, | |
| "eval_steps_per_second": 2.336, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 1.0358223915100098, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.3512, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.0665713548660278, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2628, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.9647076725959778, | |
| "eval_runtime": 5.1272, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.9864840507507324, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.1707, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.1113288402557373, | |
| "eval_runtime": 5.1295, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.8879310488700867, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.1233, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 1.2850966453552246, | |
| "eval_runtime": 5.1291, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 1.289452075958252, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.0896, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.7701128125190735, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0474, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 1.4283571243286133, | |
| "eval_runtime": 5.1313, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.7709472179412842, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.0448, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 1.488186240196228, | |
| "eval_runtime": 5.1301, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.7411565184593201, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0332, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.4506477415561676, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0198, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 1.6128920316696167, | |
| "eval_runtime": 5.1363, | |
| "eval_samples_per_second": 2.336, | |
| "eval_steps_per_second": 2.336, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "step": 34, | |
| "total_flos": 8954933126627328.0, | |
| "train_loss": 0.4336299673544572, | |
| "train_runtime": 672.6435, | |
| "train_samples_per_second": 3.419, | |
| "train_steps_per_second": 0.149 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8954933126627328.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |