| { | |
| "best_metric": 1.120043396949768, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-3_120-samples_config-2_full_auto/checkpoint-66", | |
| "epoch": 18.90909090909091, | |
| "eval_steps": 500, | |
| "global_step": 104, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.581295371055603, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.7031, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.497828871011734, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.6572, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.4343453645706177, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.6955, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 1.6712287664413452, | |
| "eval_runtime": 21.0994, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.44503721594810486, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.6969, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.472679078578949, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.6328, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.34954628348350525, | |
| "learning_rate": 4e-05, | |
| "loss": 1.6465, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.5769468545913696, | |
| "eval_runtime": 21.1031, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.29231199622154236, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.5591, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.37726345658302307, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.5185, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.24997571110725403, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.516, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 1.4843393564224243, | |
| "eval_runtime": 21.1115, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.2430959790945053, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.4624, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.2389281988143921, | |
| "learning_rate": 8e-05, | |
| "loss": 1.3985, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2073509842157364, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.374, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.3647857904434204, | |
| "eval_runtime": 21.114, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.27162283658981323, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.3274, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.30108562111854553, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 1.2447, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 1.2467561960220337, | |
| "eval_runtime": 21.1022, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.3932035267353058, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 1.2266, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.24709945917129517, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.1726, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.28725895285606384, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 1.1453, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 1.1843971014022827, | |
| "eval_runtime": 21.1049, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.16724050045013428, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 1.1348, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.21087133884429932, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 1.1125, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.15810544788837433, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 1.1195, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 1.164704442024231, | |
| "eval_runtime": 21.1083, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.14488349854946136, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.1186, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.14295697212219238, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 1.0739, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.13706757128238678, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 1.1005, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.1477621793746948, | |
| "eval_runtime": 21.1033, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.14223842322826385, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 1.067, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.16188372671604156, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 1.0586, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 1.1370995044708252, | |
| "eval_runtime": 21.1021, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.1443658322095871, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.0771, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.1483730673789978, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 1.0249, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.14605861902236938, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 1.0782, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.1279171705245972, | |
| "eval_runtime": 21.1035, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.1600651890039444, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 1.0138, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.17749443650245667, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 1.0125, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.16405703127384186, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 1.0254, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 1.1231459379196167, | |
| "eval_runtime": 21.1063, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.17835721373558044, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 1.0085, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.16730767488479614, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 1.0152, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.2164454162120819, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.9925, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.120043396949768, | |
| "eval_runtime": 21.1069, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.19749857485294342, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.9818, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.21602660417556763, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.9598, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 1.1200594902038574, | |
| "eval_runtime": 21.1138, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.229014053940773, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.9517, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.2029261440038681, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.9949, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.22543363273143768, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.9553, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.125894546508789, | |
| "eval_runtime": 21.1025, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.181818181818182, | |
| "grad_norm": 0.2360914945602417, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.9018, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 14.545454545454545, | |
| "grad_norm": 0.25790664553642273, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.9588, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "grad_norm": 0.33594781160354614, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.8505, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.909090909090908, | |
| "eval_loss": 1.1371972560882568, | |
| "eval_runtime": 21.1081, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 15.272727272727273, | |
| "grad_norm": 0.38251662254333496, | |
| "learning_rate": 8.397206521307584e-05, | |
| "loss": 0.9056, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 15.636363636363637, | |
| "grad_norm": 0.4161888062953949, | |
| "learning_rate": 8.293447300593402e-05, | |
| "loss": 0.8601, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.3428288698196411, | |
| "learning_rate": 8.18711994874345e-05, | |
| "loss": 0.865, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.1482844352722168, | |
| "eval_runtime": 21.1041, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.363636363636363, | |
| "grad_norm": 0.29455941915512085, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.9071, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.727272727272727, | |
| "grad_norm": 0.4821699261665344, | |
| "learning_rate": 7.967094433018508e-05, | |
| "loss": 0.8046, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.90909090909091, | |
| "eval_loss": 1.1631736755371094, | |
| "eval_runtime": 21.1086, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 17.09090909090909, | |
| "grad_norm": 0.4916638731956482, | |
| "learning_rate": 7.85356783842216e-05, | |
| "loss": 0.7779, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 17.454545454545453, | |
| "grad_norm": 0.45283034443855286, | |
| "learning_rate": 7.737816117462752e-05, | |
| "loss": 0.7811, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 17.818181818181817, | |
| "grad_norm": 0.4605153501033783, | |
| "learning_rate": 7.619929529850397e-05, | |
| "loss": 0.7313, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 1.1917126178741455, | |
| "eval_runtime": 21.1029, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 18.181818181818183, | |
| "grad_norm": 0.5665600895881653, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.7806, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 18.545454545454547, | |
| "grad_norm": 1.4343844652175903, | |
| "learning_rate": 7.378121045351378e-05, | |
| "loss": 0.7148, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "grad_norm": 0.7419711947441101, | |
| "learning_rate": 7.254387703447154e-05, | |
| "loss": 0.7125, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "eval_loss": 1.2146124839782715, | |
| "eval_runtime": 21.105, | |
| "eval_samples_per_second": 1.137, | |
| "eval_steps_per_second": 1.137, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 18.90909090909091, | |
| "step": 104, | |
| "total_flos": 1.3173398632923136e+17, | |
| "train_loss": 1.1138932842474718, | |
| "train_runtime": 4848.2529, | |
| "train_samples_per_second": 0.908, | |
| "train_steps_per_second": 0.052 | |
| } | |
| ], | |
| "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": 1.3173398632923136e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |