| { | |
| "best_metric": 0.8733854293823242, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-1_full/checkpoint-34", | |
| "epoch": 12.869565217391305, | |
| "eval_steps": 500, | |
| "global_step": 74, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 0.977586030960083, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.2526, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.1381486654281616, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2788, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.9461638331413269, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.2252, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.1980607509613037, | |
| "eval_runtime": 13.8108, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.861394464969635, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.2099, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.608093798160553, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.1723, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.4019879400730133, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1411, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.0843513011932373, | |
| "eval_runtime": 13.8053, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.4438973367214203, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.1031, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.32766103744506836, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.0271, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.3773520290851593, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.0554, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.9996404647827148, | |
| "eval_runtime": 13.8037, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.4237472116947174, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.9587, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.5382972359657288, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9361, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.42597174644470215, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.8972, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9073739051818848, | |
| "eval_runtime": 13.8023, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.3177820146083832, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8962, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.2999187707901001, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.8616, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.30600398778915405, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.8159, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8847095370292664, | |
| "eval_runtime": 13.801, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.2666756212711334, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7828, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.2639431655406952, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.8166, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.25138649344444275, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.7851, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8733854293823242, | |
| "eval_runtime": 13.808, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.2561894953250885, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.7616, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.32361170649528503, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.7549, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.29522183537483215, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7179, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.8734186291694641, | |
| "eval_runtime": 13.8042, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.32336854934692383, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.74, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.3424532413482666, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.6882, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.39140939712524414, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.6768, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.88090580701828, | |
| "eval_runtime": 13.8072, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.4047468304634094, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.6701, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.5629932284355164, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.6127, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.8959582448005676, | |
| "eval_runtime": 13.803, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.44259828329086304, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.6137, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.8049341440200806, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.5417, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.6533960700035095, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.5592, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.9449066519737244, | |
| "eval_runtime": 13.8014, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.7448545098304749, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.5538, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 1.2084859609603882, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.4633, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.9427428841590881, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.4519, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.9643160700798035, | |
| "eval_runtime": 13.8034, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.7851306200027466, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.4258, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 1.991536021232605, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.4546, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.0970418453216553, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.367, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.9817283153533936, | |
| "eval_runtime": 13.8003, | |
| "eval_samples_per_second": 0.87, | |
| "eval_steps_per_second": 0.87, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 1.0731067657470703, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.3574, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.8825430870056152, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.3049, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 1.6797748804092407, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.3385, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.1056302785873413, | |
| "eval_runtime": 13.802, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "step": 74, | |
| "total_flos": 5.893544901135565e+16, | |
| "train_loss": 0.7602536090322443, | |
| "train_runtime": 2198.2645, | |
| "train_samples_per_second": 1.046, | |
| "train_steps_per_second": 0.114 | |
| } | |
| ], | |
| "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": 5.893544901135565e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |