| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 15.0, | |
| "eval_steps": 500, | |
| "global_step": 18810, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.39872408293460926, | |
| "grad_norm": 0.7819421887397766, | |
| "learning_rate": 4.800637958532696e-05, | |
| "loss": 8.5954, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.7974481658692185, | |
| "grad_norm": 1.4368996620178223, | |
| "learning_rate": 4.6012759170653905e-05, | |
| "loss": 7.6895, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.1961722488038278, | |
| "grad_norm": 1.0771726369857788, | |
| "learning_rate": 4.401913875598087e-05, | |
| "loss": 7.5338, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.594896331738437, | |
| "grad_norm": 1.4085040092468262, | |
| "learning_rate": 4.2025518341307815e-05, | |
| "loss": 7.4062, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.9936204146730463, | |
| "grad_norm": 1.2863935232162476, | |
| "learning_rate": 4.003189792663477e-05, | |
| "loss": 7.3026, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.3923444976076556, | |
| "grad_norm": 1.343314528465271, | |
| "learning_rate": 3.8038277511961725e-05, | |
| "loss": 7.2067, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.7910685805422646, | |
| "grad_norm": 1.8292430639266968, | |
| "learning_rate": 3.604465709728867e-05, | |
| "loss": 7.1446, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 3.189792663476874, | |
| "grad_norm": 1.3264261484146118, | |
| "learning_rate": 3.4051036682615634e-05, | |
| "loss": 7.075, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.588516746411483, | |
| "grad_norm": 1.511952519416809, | |
| "learning_rate": 3.205741626794259e-05, | |
| "loss": 7.0362, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.9872408293460926, | |
| "grad_norm": 1.9909762144088745, | |
| "learning_rate": 3.0063795853269537e-05, | |
| "loss": 6.9639, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "grad_norm": 1.8488012552261353, | |
| "learning_rate": 2.8070175438596492e-05, | |
| "loss": 6.9264, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 4.784688995215311, | |
| "grad_norm": 1.4738715887069702, | |
| "learning_rate": 2.6076555023923443e-05, | |
| "loss": 6.9039, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 5.18341307814992, | |
| "grad_norm": 1.7991809844970703, | |
| "learning_rate": 2.4082934609250398e-05, | |
| "loss": 6.8603, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 5.582137161084529, | |
| "grad_norm": 1.811631441116333, | |
| "learning_rate": 2.2089314194577353e-05, | |
| "loss": 6.8354, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 5.980861244019139, | |
| "grad_norm": 1.732555627822876, | |
| "learning_rate": 2.0095693779904308e-05, | |
| "loss": 6.8048, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 6.379585326953748, | |
| "grad_norm": 1.7755110263824463, | |
| "learning_rate": 1.8102073365231263e-05, | |
| "loss": 6.7829, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 6.778309409888357, | |
| "grad_norm": 2.225064277648926, | |
| "learning_rate": 1.611244019138756e-05, | |
| "loss": 6.7659, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 7.177033492822966, | |
| "grad_norm": 1.872410774230957, | |
| "learning_rate": 1.4118819776714515e-05, | |
| "loss": 6.7374, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 7.575757575757576, | |
| "grad_norm": 1.7238253355026245, | |
| "learning_rate": 1.2125199362041468e-05, | |
| "loss": 6.7349, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 7.974481658692185, | |
| "grad_norm": 1.8047776222229004, | |
| "learning_rate": 1.0131578947368421e-05, | |
| "loss": 6.7204, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 8.373205741626794, | |
| "grad_norm": 2.0625622272491455, | |
| "learning_rate": 8.145933014354067e-06, | |
| "loss": 6.6957, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "grad_norm": 1.8150039911270142, | |
| "learning_rate": 6.152312599681021e-06, | |
| "loss": 6.7014, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 9.170653907496012, | |
| "grad_norm": 2.004660129547119, | |
| "learning_rate": 4.158692185007974e-06, | |
| "loss": 6.6855, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 9.569377990430622, | |
| "grad_norm": 2.1859395503997803, | |
| "learning_rate": 2.1650717703349284e-06, | |
| "loss": 6.6813, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 9.96810207336523, | |
| "grad_norm": 2.1167070865631104, | |
| "learning_rate": 1.7145135566188198e-07, | |
| "loss": 6.6833, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 12540, | |
| "total_flos": 5910924298141440.0, | |
| "train_loss": 7.017849675472083, | |
| "train_runtime": 3142.2268, | |
| "train_samples_per_second": 510.596, | |
| "train_steps_per_second": 3.991 | |
| }, | |
| { | |
| "epoch": 10.36682615629984, | |
| "grad_norm": 2.343043327331543, | |
| "learning_rate": 4.87772461456672e-05, | |
| "loss": 6.6945, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 10.76555023923445, | |
| "grad_norm": 2.0774621963500977, | |
| "learning_rate": 4.7448165869218504e-05, | |
| "loss": 6.6465, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 11.164274322169058, | |
| "grad_norm": 2.5551674365997314, | |
| "learning_rate": 4.611908559276981e-05, | |
| "loss": 6.6086, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 11.562998405103668, | |
| "grad_norm": 2.335573196411133, | |
| "learning_rate": 4.479000531632111e-05, | |
| "loss": 6.5677, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 11.961722488038278, | |
| "grad_norm": 2.7488040924072266, | |
| "learning_rate": 4.346092503987241e-05, | |
| "loss": 6.5439, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 12.360446570972886, | |
| "grad_norm": 2.1827917098999023, | |
| "learning_rate": 4.213184476342371e-05, | |
| "loss": 6.4975, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 12.759170653907496, | |
| "grad_norm": 2.1983468532562256, | |
| "learning_rate": 4.080276448697501e-05, | |
| "loss": 6.4953, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 13.157894736842104, | |
| "grad_norm": 2.4948813915252686, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 6.4546, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 13.556618819776714, | |
| "grad_norm": 2.0236194133758545, | |
| "learning_rate": 3.814460393407762e-05, | |
| "loss": 6.432, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 13.955342902711324, | |
| "grad_norm": 2.3203213214874268, | |
| "learning_rate": 3.681552365762892e-05, | |
| "loss": 6.4038, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 14.354066985645932, | |
| "grad_norm": 2.347102165222168, | |
| "learning_rate": 3.5486443381180226e-05, | |
| "loss": 6.3872, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 14.752791068580542, | |
| "grad_norm": 2.5316317081451416, | |
| "learning_rate": 3.415736310473153e-05, | |
| "loss": 6.3687, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "step": 18810, | |
| "total_flos": 8866386447212160.0, | |
| "train_loss": 2.1665261722384206, | |
| "train_runtime": 1599.9695, | |
| "train_samples_per_second": 1504.163, | |
| "train_steps_per_second": 11.756 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 18810, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 15, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8866386447212160.0, | |
| "train_batch_size": 128, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |