| { | |
| "best_metric": 0.9256513714790344, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_60-samples_config-1_full/checkpoint-28", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 69, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 2.6855738162994385, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.0455, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 3.0714163780212402, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.237, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.6781492233276367, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.1428, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.8871607780456543, | |
| "eval_runtime": 4.8512, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 2.191319227218628, | |
| "learning_rate": 2.4e-05, | |
| "loss": 2.0036, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 2.064764976501465, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.8981, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 1.388030767440796, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7313, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.4963082075119019, | |
| "eval_runtime": 4.843, | |
| "eval_samples_per_second": 2.478, | |
| "eval_steps_per_second": 2.478, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 1.3937901258468628, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.5883, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 1.6199463605880737, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.427, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.3204479217529297, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.2743, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 1.1005104780197144, | |
| "eval_runtime": 4.851, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 1.0653191804885864, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.1306, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.9779132008552551, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9714, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.9297244548797607, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.9489, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9685582518577576, | |
| "eval_runtime": 4.8406, | |
| "eval_samples_per_second": 2.479, | |
| "eval_steps_per_second": 2.479, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.9527274966239929, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8866, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.8115637302398682, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.8781, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.813915491104126, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.7984, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.9256513714790344, | |
| "eval_runtime": 4.8495, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.8284777402877808, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7507, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.9932141304016113, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.7032, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 1.071712851524353, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.6358, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.9723448157310486, | |
| "eval_runtime": 4.8499, | |
| "eval_samples_per_second": 2.474, | |
| "eval_steps_per_second": 2.474, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 1.0750497579574585, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.5489, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 1.2155307531356812, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.519, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.2024857997894287, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.4722, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.0250778198242188, | |
| "eval_runtime": 4.8457, | |
| "eval_samples_per_second": 2.476, | |
| "eval_steps_per_second": 2.476, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 1.261959433555603, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.4044, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 1.2990663051605225, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.3495, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.43495512008667, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.3258, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.124754548072815, | |
| "eval_runtime": 4.8437, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 2.477, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 1.6028568744659424, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.2656, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 1.2100633382797241, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2366, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 1.241849422454834, | |
| "eval_runtime": 4.8536, | |
| "eval_samples_per_second": 2.472, | |
| "eval_steps_per_second": 2.472, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 1.2336710691452026, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.2002, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 1.5955331325531006, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.1805, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 1.4076216220855713, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.1887, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 1.300315499305725, | |
| "eval_runtime": 4.8422, | |
| "eval_samples_per_second": 2.478, | |
| "eval_steps_per_second": 2.478, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 1.1033974885940552, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.1585, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 1.1297698020935059, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.1436, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 1.5109285116195679, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.1497, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 1.3586145639419556, | |
| "eval_runtime": 4.8487, | |
| "eval_samples_per_second": 2.475, | |
| "eval_steps_per_second": 2.475, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.8872708082199097, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.1372, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.9570254683494568, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.1194, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.7731051445007324, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.1313, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.4446760416030884, | |
| "eval_runtime": 4.8424, | |
| "eval_samples_per_second": 2.478, | |
| "eval_steps_per_second": 2.478, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 69, | |
| "total_flos": 8077008546299904.0, | |
| "train_loss": 0.768153426979763, | |
| "train_runtime": 647.6155, | |
| "train_samples_per_second": 3.551, | |
| "train_steps_per_second": 0.386 | |
| } | |
| ], | |
| "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": 8077008546299904.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |