| { | |
| "best_metric": 1.1440507173538208, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_60-samples_config-1/checkpoint-23", | |
| "epoch": 10.956521739130435, | |
| "eval_steps": 500, | |
| "global_step": 63, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 5.991142272949219, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.2123, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 6.2022881507873535, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.3365, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 5.94657564163208, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 2.3685, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.9770396947860718, | |
| "eval_runtime": 4.9096, | |
| "eval_samples_per_second": 2.444, | |
| "eval_steps_per_second": 2.444, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 5.112995624542236, | |
| "learning_rate": 2.4e-05, | |
| "loss": 2.162, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 4.547484874725342, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.9474, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 3.081784248352051, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5814, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.3894500732421875, | |
| "eval_runtime": 4.9463, | |
| "eval_samples_per_second": 2.426, | |
| "eval_steps_per_second": 2.426, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 2.3743317127227783, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.354, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 1.6964478492736816, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.2551, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.6185894012451172, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.2096, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 1.2663203477859497, | |
| "eval_runtime": 4.8945, | |
| "eval_samples_per_second": 2.452, | |
| "eval_steps_per_second": 2.452, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 1.5679247379302979, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.1433, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.7541425228118896, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9958, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 1.479678988456726, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.0552, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.1440507173538208, | |
| "eval_runtime": 4.8958, | |
| "eval_samples_per_second": 2.451, | |
| "eval_steps_per_second": 2.451, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 1.4151923656463623, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.9004, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 1.4100896120071411, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.7865, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 1.4641093015670776, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.7016, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 1.2034224271774292, | |
| "eval_runtime": 4.9023, | |
| "eval_samples_per_second": 2.448, | |
| "eval_steps_per_second": 2.448, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 2.018409490585327, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6294, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 1.8171939849853516, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.4796, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 2.3160622119903564, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.3712, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.5233873128890991, | |
| "eval_runtime": 4.9388, | |
| "eval_samples_per_second": 2.43, | |
| "eval_steps_per_second": 2.43, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 2.1395063400268555, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.2496, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 3.050215721130371, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.1929, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 3.4642417430877686, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.2113, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.7754243612289429, | |
| "eval_runtime": 4.9005, | |
| "eval_samples_per_second": 2.449, | |
| "eval_steps_per_second": 2.449, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 1.621311902999878, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.0883, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 2.752639055252075, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.0951, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.989722490310669, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.1125, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.8850336074829102, | |
| "eval_runtime": 4.8946, | |
| "eval_samples_per_second": 2.452, | |
| "eval_steps_per_second": 2.452, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 3.1502797603607178, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.0401, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 1.9465205669403076, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0539, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 2.033078908920288, | |
| "eval_runtime": 4.845, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 2.477, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 1.212947130203247, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.0254, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 2.585963487625122, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.0464, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 2.3054916858673096, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.0327, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 2.0972988605499268, | |
| "eval_runtime": 4.8449, | |
| "eval_samples_per_second": 2.477, | |
| "eval_steps_per_second": 2.477, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 1.6045154333114624, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.0346, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.9054758548736572, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0141, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.3107316195964813, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.0269, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 2.025130033493042, | |
| "eval_runtime": 4.8408, | |
| "eval_samples_per_second": 2.479, | |
| "eval_steps_per_second": 2.479, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "step": 63, | |
| "total_flos": 7403924491337728.0, | |
| "train_loss": 0.7132057910932909, | |
| "train_runtime": 598.2674, | |
| "train_samples_per_second": 3.844, | |
| "train_steps_per_second": 0.418 | |
| } | |
| ], | |
| "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": 7403924491337728.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |