| { | |
| "best_metric": 0.8653757572174072, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-1_full_auto/checkpoint-40", | |
| "epoch": 13.91304347826087, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 0.9382457733154297, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.2528, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.0488719940185547, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2409, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.9187086820602417, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.2067, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.1800121068954468, | |
| "eval_runtime": 13.9452, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.8312786221504211, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.1927, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.6022394299507141, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.1548, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.4049568772315979, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1313, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.0718904733657837, | |
| "eval_runtime": 13.9427, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.41703853011131287, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.0922, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.31836652755737305, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.0217, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.383449524641037, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.0437, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.9889429211616516, | |
| "eval_runtime": 13.9431, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.44971293210983276, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.9513, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.5273430347442627, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9261, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.3666059970855713, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.8887, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8997397422790527, | |
| "eval_runtime": 13.9445, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.3411215841770172, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8939, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.3052791655063629, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.8506, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.31273409724235535, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.8094, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8759673237800598, | |
| "eval_runtime": 13.9475, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.3146740198135376, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7763, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.2547754943370819, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.8117, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.2573046088218689, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.7792, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8671879172325134, | |
| "eval_runtime": 13.9437, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.2736125588417053, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.759, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.352104127407074, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.7449, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.307273268699646, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7139, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.8653757572174072, | |
| "eval_runtime": 13.9454, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.31662505865097046, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.7353, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.37183788418769836, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.6844, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.4128043055534363, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.666, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8718865513801575, | |
| "eval_runtime": 13.9433, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.4202701151371002, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.6641, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.6468128561973572, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.6045, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.8942766785621643, | |
| "eval_runtime": 13.9521, | |
| "eval_samples_per_second": 0.86, | |
| "eval_steps_per_second": 0.86, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.45054003596305847, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.6049, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.683666467666626, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.5372, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.8305115699768066, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.5506, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.9270607829093933, | |
| "eval_runtime": 13.9397, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.6290498971939087, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.5466, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.7966501712799072, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.461, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.9955141544342041, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.4489, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.9575808048248291, | |
| "eval_runtime": 13.9386, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.8269140720367432, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.4159, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 1.2511094808578491, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.4437, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.3201357126235962, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.3597, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.0070308446884155, | |
| "eval_runtime": 13.9404, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 1.0811384916305542, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.342, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.9966951012611389, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.2959, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 1.5296682119369507, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.3326, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.080146074295044, | |
| "eval_runtime": 13.9423, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.9342603087425232, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.2861, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 1.2436318397521973, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.2663, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 1.3493071794509888, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.2557, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.2003072500228882, | |
| "eval_runtime": 13.9438, | |
| "eval_samples_per_second": 0.861, | |
| "eval_steps_per_second": 0.861, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "step": 80, | |
| "total_flos": 6.407243199571558e+16, | |
| "train_loss": 0.7138225329108536, | |
| "train_runtime": 2386.9247, | |
| "train_samples_per_second": 0.964, | |
| "train_steps_per_second": 0.105 | |
| } | |
| ], | |
| "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": 6.407243199571558e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |