| { | |
| "best_metric": 0.789516031742096, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_60-samples_config-1_auto/checkpoint-17", | |
| "epoch": 9.91304347826087, | |
| "eval_steps": 500, | |
| "global_step": 57, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 2.7129995822906494, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9538, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 3.2067904472351074, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.327, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.6511616706848145, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.0663, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.0319761037826538, | |
| "eval_runtime": 5.1301, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 2.33830189704895, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.0422, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 2.130410671234131, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.8733, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 1.8173961639404297, | |
| "learning_rate": 4e-05, | |
| "loss": 0.887, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 0.8523615002632141, | |
| "eval_runtime": 5.1304, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 1.7136945724487305, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.8822, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 1.4543921947479248, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.7376, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.4940763711929321, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.7166, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.789516031742096, | |
| "eval_runtime": 5.1334, | |
| "eval_samples_per_second": 2.338, | |
| "eval_steps_per_second": 2.338, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 1.4083375930786133, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.6811, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.3556171655654907, | |
| "learning_rate": 8e-05, | |
| "loss": 0.554, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 1.487412929534912, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.5246, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8278915286064148, | |
| "eval_runtime": 5.131, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 1.6211680173873901, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.5259, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 1.5383013486862183, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.3497, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 1.566432237625122, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.3244, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.876640260219574, | |
| "eval_runtime": 5.1326, | |
| "eval_samples_per_second": 2.338, | |
| "eval_steps_per_second": 2.338, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 1.3085135221481323, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.236, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 1.6528167724609375, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.1684, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 2.083045244216919, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.2036, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.171944499015808, | |
| "eval_runtime": 5.1345, | |
| "eval_samples_per_second": 2.337, | |
| "eval_steps_per_second": 2.337, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 1.4387879371643066, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.0922, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 2.0251829624176025, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.1098, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.2941664457321167, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.069, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.2153328657150269, | |
| "eval_runtime": 5.1313, | |
| "eval_samples_per_second": 2.339, | |
| "eval_steps_per_second": 2.339, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.98664391040802, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.0581, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.9605448246002197, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.0472, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.2975889444351196, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.0557, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.349020004272461, | |
| "eval_runtime": 5.1322, | |
| "eval_samples_per_second": 2.338, | |
| "eval_steps_per_second": 2.338, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 1.239974856376648, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.0337, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.4344079792499542, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0301, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 1.4460724592208862, | |
| "eval_runtime": 5.1316, | |
| "eval_samples_per_second": 2.338, | |
| "eval_steps_per_second": 2.338, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.8744381070137024, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.0317, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.6762619614601135, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.0208, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 1.8280589580535889, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.0269, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 1.5282880067825317, | |
| "eval_runtime": 5.1361, | |
| "eval_samples_per_second": 2.336, | |
| "eval_steps_per_second": 2.336, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "step": 57, | |
| "total_flos": 7462444286869504.0, | |
| "train_loss": 0.40365017078032617, | |
| "train_runtime": 561.8805, | |
| "train_samples_per_second": 4.093, | |
| "train_steps_per_second": 0.445 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 250, | |
| "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": 7462444286869504.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |