| { | |
| "best_metric": 0.13464748859405518, | |
| "best_model_checkpoint": "./lora-out/checkpoint-204", | |
| "epoch": 14.442477876106194, | |
| "eval_steps": 4, | |
| "global_step": 204, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 0.3006412982940674, | |
| "eval_runtime": 54.7333, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 0.300335168838501, | |
| "eval_runtime": 54.8113, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 1e-05, | |
| "loss": 0.3024, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 0.2993900179862976, | |
| "eval_runtime": 54.782, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_loss": 0.29816487431526184, | |
| "eval_runtime": 54.8049, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 2e-05, | |
| "loss": 0.3035, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_loss": 0.29595255851745605, | |
| "eval_runtime": 54.7885, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 0.2939557135105133, | |
| "eval_runtime": 54.7999, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "eval_loss": 0.29013773798942566, | |
| "eval_runtime": 54.7805, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "learning_rate": 3e-05, | |
| "loss": 0.2959, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "eval_loss": 0.28251081705093384, | |
| "eval_runtime": 54.7706, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 2.55, | |
| "eval_loss": 0.2771329879760742, | |
| "eval_runtime": 54.7818, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "learning_rate": 4e-05, | |
| "loss": 0.284, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.83, | |
| "eval_loss": 0.27146488428115845, | |
| "eval_runtime": 54.803, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "eval_loss": 0.26464152336120605, | |
| "eval_runtime": 54.8467, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "eval_loss": 0.25653430819511414, | |
| "eval_runtime": 54.8327, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 5e-05, | |
| "loss": 0.263, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "eval_loss": 0.24627122282981873, | |
| "eval_runtime": 54.813, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 3.96, | |
| "eval_loss": 0.23474617302417755, | |
| "eval_runtime": 54.7901, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "learning_rate": 6e-05, | |
| "loss": 0.241, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.25, | |
| "eval_loss": 0.2220366895198822, | |
| "eval_runtime": 54.7983, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 4.53, | |
| "eval_loss": 0.20926769077777863, | |
| "eval_runtime": 54.7403, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 4.81, | |
| "eval_loss": 0.19629451632499695, | |
| "eval_runtime": 54.7525, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "learning_rate": 7e-05, | |
| "loss": 0.2101, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 5.1, | |
| "eval_loss": 0.18524658679962158, | |
| "eval_runtime": 54.7303, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "eval_loss": 0.17731742560863495, | |
| "eval_runtime": 54.7552, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "learning_rate": 8e-05, | |
| "loss": 0.1788, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.66, | |
| "eval_loss": 0.16993452608585358, | |
| "eval_runtime": 54.729, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 5.95, | |
| "eval_loss": 0.164781853556633, | |
| "eval_runtime": 54.741, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 6.23, | |
| "eval_loss": 0.16103117167949677, | |
| "eval_runtime": 54.7837, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 6.37, | |
| "learning_rate": 9e-05, | |
| "loss": 0.1615, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 6.51, | |
| "eval_loss": 0.15781742334365845, | |
| "eval_runtime": 54.7138, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "eval_loss": 0.15516981482505798, | |
| "eval_runtime": 54.7516, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "learning_rate": 0.0001, | |
| "loss": 0.1533, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.08, | |
| "eval_loss": 0.15261690318584442, | |
| "eval_runtime": 54.6891, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "eval_loss": 0.15066812932491302, | |
| "eval_runtime": 54.6884, | |
| "eval_samples_per_second": 0.914, | |
| "eval_steps_per_second": 0.238, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 7.65, | |
| "eval_loss": 0.14893724024295807, | |
| "eval_runtime": 54.6275, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 7.79, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.1463, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 7.93, | |
| "eval_loss": 0.14742153882980347, | |
| "eval_runtime": 54.6174, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 8.21, | |
| "eval_loss": 0.14575307071208954, | |
| "eval_runtime": 54.6366, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 8.181818181818183e-05, | |
| "loss": 0.1399, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "eval_loss": 0.14450186491012573, | |
| "eval_runtime": 54.6303, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "eval_loss": 0.1431863009929657, | |
| "eval_runtime": 54.6358, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 9.06, | |
| "eval_loss": 0.1424635797739029, | |
| "eval_runtime": 54.6449, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 128 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.1357, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 9.35, | |
| "eval_loss": 0.14175941050052643, | |
| "eval_runtime": 54.6307, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 9.63, | |
| "eval_loss": 0.14105737209320068, | |
| "eval_runtime": 54.6121, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "learning_rate": 6.363636363636364e-05, | |
| "loss": 0.1322, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 9.91, | |
| "eval_loss": 0.14027251303195953, | |
| "eval_runtime": 54.6594, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 10.19, | |
| "eval_loss": 0.13963991403579712, | |
| "eval_runtime": 54.6378, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 10.48, | |
| "eval_loss": 0.138994500041008, | |
| "eval_runtime": 54.6115, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 10.62, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.1355, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 10.76, | |
| "eval_loss": 0.13857363164424896, | |
| "eval_runtime": 54.6622, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 11.04, | |
| "eval_loss": 0.13809233903884888, | |
| "eval_runtime": 54.7824, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 11.33, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.1216, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 11.33, | |
| "eval_loss": 0.137764573097229, | |
| "eval_runtime": 54.6049, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "eval_loss": 0.1369408369064331, | |
| "eval_runtime": 54.6188, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 11.89, | |
| "eval_loss": 0.13684938848018646, | |
| "eval_runtime": 54.5949, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 12.04, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.1265, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 12.18, | |
| "eval_loss": 0.1366124004125595, | |
| "eval_runtime": 54.5928, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 12.46, | |
| "eval_loss": 0.1361435353755951, | |
| "eval_runtime": 54.6703, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "learning_rate": 2.7272727272727273e-05, | |
| "loss": 0.127, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 12.74, | |
| "eval_loss": 0.13553684949874878, | |
| "eval_runtime": 54.6232, | |
| "eval_samples_per_second": 0.915, | |
| "eval_steps_per_second": 0.238, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 13.03, | |
| "eval_loss": 0.13531364500522614, | |
| "eval_runtime": 54.7712, | |
| "eval_samples_per_second": 0.913, | |
| "eval_steps_per_second": 0.237, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 13.31, | |
| "eval_loss": 0.1353050172328949, | |
| "eval_runtime": 54.8052, | |
| "eval_samples_per_second": 0.912, | |
| "eval_steps_per_second": 0.237, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 13.45, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.1233, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 13.59, | |
| "eval_loss": 0.1348796784877777, | |
| "eval_runtime": 54.5607, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "eval_loss": 0.13486500084400177, | |
| "eval_runtime": 54.5618, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "learning_rate": 9.090909090909091e-06, | |
| "loss": 0.1189, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 14.16, | |
| "eval_loss": 0.13471217453479767, | |
| "eval_runtime": 54.6065, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 14.44, | |
| "eval_loss": 0.13464748859405518, | |
| "eval_runtime": 54.5803, | |
| "eval_samples_per_second": 0.916, | |
| "eval_steps_per_second": 0.238, | |
| "step": 204 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 12, | |
| "total_flos": 6.711268353280573e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |