| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.2284122562674096, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.055710306406685235, | |
| "grad_norm": 13.33180221602099, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.7002, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11142061281337047, | |
| "grad_norm": 1.6650732524504632, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.2272, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.1671309192200557, | |
| "grad_norm": 1.3298246517786452, | |
| "learning_rate": 5.555555555555557e-06, | |
| "loss": 0.1798, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.22284122562674094, | |
| "grad_norm": 1.5154839686376809, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1663, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.2785515320334262, | |
| "grad_norm": 1.4493281388620185, | |
| "learning_rate": 9.25925925925926e-06, | |
| "loss": 0.1698, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3342618384401114, | |
| "grad_norm": 1.4507180686803252, | |
| "learning_rate": 9.99619291237835e-06, | |
| "loss": 0.1547, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.38997214484679665, | |
| "grad_norm": 1.408311153993914, | |
| "learning_rate": 9.972948368899302e-06, | |
| "loss": 0.1573, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4456824512534819, | |
| "grad_norm": 1.411285032932152, | |
| "learning_rate": 9.928672515804434e-06, | |
| "loss": 0.1651, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5013927576601671, | |
| "grad_norm": 1.5921373458809418, | |
| "learning_rate": 9.863552602006435e-06, | |
| "loss": 0.1564, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5571030640668524, | |
| "grad_norm": 1.1123208209611752, | |
| "learning_rate": 9.777864028930705e-06, | |
| "loss": 0.1599, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5571030640668524, | |
| "eval_loss": 0.14965404570102692, | |
| "eval_runtime": 14.3523, | |
| "eval_samples_per_second": 42.084, | |
| "eval_steps_per_second": 1.324, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6128133704735376, | |
| "grad_norm": 1.3165488984028846, | |
| "learning_rate": 9.671969185803357e-06, | |
| "loss": 0.1539, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.6685236768802229, | |
| "grad_norm": 1.2598963561948195, | |
| "learning_rate": 9.546315917055362e-06, | |
| "loss": 0.158, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.724233983286908, | |
| "grad_norm": 1.2565192353667611, | |
| "learning_rate": 9.401435628324436e-06, | |
| "loss": 0.172, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.7799442896935933, | |
| "grad_norm": 1.3015497330703336, | |
| "learning_rate": 9.237941039064606e-06, | |
| "loss": 0.1604, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8356545961002786, | |
| "grad_norm": 1.3910423306638404, | |
| "learning_rate": 9.056523591268064e-06, | |
| "loss": 0.1552, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8913649025069638, | |
| "grad_norm": 1.1494310395657203, | |
| "learning_rate": 8.85795052525811e-06, | |
| "loss": 0.1623, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.947075208913649, | |
| "grad_norm": 1.1057721484907344, | |
| "learning_rate": 8.643061634920146e-06, | |
| "loss": 0.159, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0027855153203342, | |
| "grad_norm": 1.4188078858125839, | |
| "learning_rate": 8.412765716093273e-06, | |
| "loss": 0.1627, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0584958217270195, | |
| "grad_norm": 0.7623378341371474, | |
| "learning_rate": 8.168036723142743e-06, | |
| "loss": 0.0859, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1142061281337048, | |
| "grad_norm": 1.0235514533508279, | |
| "learning_rate": 7.909909649967735e-06, | |
| "loss": 0.0851, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.1142061281337048, | |
| "eval_loss": 0.16079136729240417, | |
| "eval_runtime": 13.581, | |
| "eval_samples_per_second": 44.474, | |
| "eval_steps_per_second": 1.399, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.16991643454039, | |
| "grad_norm": 0.8096480773426198, | |
| "learning_rate": 7.639476152864163e-06, | |
| "loss": 0.0793, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.2256267409470751, | |
| "grad_norm": 0.8281128819137961, | |
| "learning_rate": 7.35787993375414e-06, | |
| "loss": 0.0812, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.2813370473537604, | |
| "grad_norm": 0.9370312703563903, | |
| "learning_rate": 7.066311903307033e-06, | |
| "loss": 0.0791, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.3370473537604457, | |
| "grad_norm": 0.9053649627789484, | |
| "learning_rate": 6.76600514440799e-06, | |
| "loss": 0.084, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.392757660167131, | |
| "grad_norm": 0.9028081686357128, | |
| "learning_rate": 6.458229697274125e-06, | |
| "loss": 0.0872, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.448467966573816, | |
| "grad_norm": 0.8100454278966028, | |
| "learning_rate": 6.144287188272867e-06, | |
| "loss": 0.0845, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.5041782729805013, | |
| "grad_norm": 1.1552403857017284, | |
| "learning_rate": 5.825505325157962e-06, | |
| "loss": 0.0837, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.5598885793871866, | |
| "grad_norm": 0.8948418601771239, | |
| "learning_rate": 5.503232282003569e-06, | |
| "loss": 0.0828, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.615598885793872, | |
| "grad_norm": 0.8307633796010723, | |
| "learning_rate": 5.178830997583353e-06, | |
| "loss": 0.0839, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.6713091922005572, | |
| "grad_norm": 0.9993028011882762, | |
| "learning_rate": 4.853673411307564e-06, | |
| "loss": 0.0793, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.6713091922005572, | |
| "eval_loss": 0.15588445961475372, | |
| "eval_runtime": 13.5881, | |
| "eval_samples_per_second": 44.451, | |
| "eval_steps_per_second": 1.398, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.7270194986072425, | |
| "grad_norm": 1.029481086800917, | |
| "learning_rate": 4.529134661095114e-06, | |
| "loss": 0.0755, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.7827298050139275, | |
| "grad_norm": 0.8067395371226609, | |
| "learning_rate": 4.206587267718743e-06, | |
| "loss": 0.0827, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.8384401114206128, | |
| "grad_norm": 0.9301538402093087, | |
| "learning_rate": 3.887395330218429e-06, | |
| "loss": 0.0825, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.894150417827298, | |
| "grad_norm": 0.8980361744773078, | |
| "learning_rate": 3.5729087569315284e-06, | |
| "loss": 0.0817, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.9498607242339832, | |
| "grad_norm": 0.9129952399332847, | |
| "learning_rate": 3.26445755653744e-06, | |
| "loss": 0.0856, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.0055710306406684, | |
| "grad_norm": 0.5738192618476439, | |
| "learning_rate": 2.963346213260737e-06, | |
| "loss": 0.0829, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.0612813370473537, | |
| "grad_norm": 0.6386795168831956, | |
| "learning_rate": 2.6708481700208954e-06, | |
| "loss": 0.0314, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.116991643454039, | |
| "grad_norm": 0.8423736663221459, | |
| "learning_rate": 2.3882004428601213e-06, | |
| "loss": 0.0273, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.1727019498607243, | |
| "grad_norm": 0.696882651835955, | |
| "learning_rate": 2.1165983894256647e-06, | |
| "loss": 0.0301, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.2284122562674096, | |
| "grad_norm": 0.9565822826782887, | |
| "learning_rate": 1.8571906536314233e-06, | |
| "loss": 0.0275, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.2284122562674096, | |
| "eval_loss": 0.19155950844287872, | |
| "eval_runtime": 13.6617, | |
| "eval_samples_per_second": 44.211, | |
| "eval_steps_per_second": 1.391, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 537, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 67047933214720.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |