| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.3515948819607138, | |
| "eval_steps": 500, | |
| "global_step": 15000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04505316273202379, | |
| "grad_norm": 2220.93115234375, | |
| "learning_rate": 1.25e-05, | |
| "loss": 517.9313, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.09010632546404758, | |
| "grad_norm": 3532.487060546875, | |
| "learning_rate": 2.5e-05, | |
| "loss": 447.0173, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13515948819607138, | |
| "grad_norm": 3076.73876953125, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 403.1765, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.18021265092809516, | |
| "grad_norm": 2786.041015625, | |
| "learning_rate": 5e-05, | |
| "loss": 371.2528, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.22526581366011894, | |
| "grad_norm": 2246.498046875, | |
| "learning_rate": 4.990486745229364e-05, | |
| "loss": 383.6465, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.27031897639214275, | |
| "grad_norm": 0.0, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 368.1067, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.3153721391241665, | |
| "grad_norm": 3346.040771484375, | |
| "learning_rate": 4.914814565722671e-05, | |
| "loss": 354.079, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.3604253018561903, | |
| "grad_norm": 3417.40185546875, | |
| "learning_rate": 4.849231551964771e-05, | |
| "loss": 351.5845, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.40547846458821407, | |
| "grad_norm": 2524.94970703125, | |
| "learning_rate": 4.765769467591625e-05, | |
| "loss": 343.0565, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.4505316273202379, | |
| "grad_norm": 2467.03076171875, | |
| "learning_rate": 4.665063509461097e-05, | |
| "loss": 325.0454, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.4955847900522617, | |
| "grad_norm": 2483.644287109375, | |
| "learning_rate": 4.54788011072248e-05, | |
| "loss": 332.3088, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.5406379527842855, | |
| "grad_norm": 3229.221923828125, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 329.884, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.5856911155163093, | |
| "grad_norm": 2195.654541015625, | |
| "learning_rate": 4.267766952966369e-05, | |
| "loss": 334.0953, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.630744278248333, | |
| "grad_norm": 2431.37744140625, | |
| "learning_rate": 4.1069690242163484e-05, | |
| "loss": 314.5535, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.6757974409803568, | |
| "grad_norm": 2344.476318359375, | |
| "learning_rate": 3.933941090877615e-05, | |
| "loss": 322.5146, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.7208506037123806, | |
| "grad_norm": 2241.12255859375, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 314.8933, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.7659037664444044, | |
| "grad_norm": 0.0, | |
| "learning_rate": 3.556545654351749e-05, | |
| "loss": 311.8407, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8109569291764281, | |
| "grad_norm": 2344.2255859375, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 303.5129, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.856010091908452, | |
| "grad_norm": 2038.534912109375, | |
| "learning_rate": 3.147047612756302e-05, | |
| "loss": 303.3366, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.9010632546404758, | |
| "grad_norm": 1972.404296875, | |
| "learning_rate": 2.9341204441673266e-05, | |
| "loss": 310.6575, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.9461164173724995, | |
| "grad_norm": 0.0, | |
| "learning_rate": 2.717889356869146e-05, | |
| "loss": 291.44, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.9911695801045234, | |
| "grad_norm": 3448.444091796875, | |
| "learning_rate": 2.5e-05, | |
| "loss": 303.3335, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 342.4247741699219, | |
| "eval_runtime": 247.7832, | |
| "eval_samples_per_second": 39.813, | |
| "eval_steps_per_second": 4.98, | |
| "step": 11098 | |
| }, | |
| { | |
| "epoch": 1.0362227428365471, | |
| "grad_norm": 4117.76806640625, | |
| "learning_rate": 2.2821106431308544e-05, | |
| "loss": 283.8551, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.081275905568571, | |
| "grad_norm": 2747.44287109375, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 268.1614, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1263290683005946, | |
| "grad_norm": 2435.322265625, | |
| "learning_rate": 1.852952387243698e-05, | |
| "loss": 270.2856, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.1713822310326185, | |
| "grad_norm": 3295.06005859375, | |
| "learning_rate": 1.6449496416858284e-05, | |
| "loss": 258.1864, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.2164353937646424, | |
| "grad_norm": 4224.72412109375, | |
| "learning_rate": 1.443454345648252e-05, | |
| "loss": 264.1837, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.261488556496666, | |
| "grad_norm": 2986.144775390625, | |
| "learning_rate": 1.2500000000000006e-05, | |
| "loss": 261.3931, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.30654171922869, | |
| "grad_norm": 2386.795166015625, | |
| "learning_rate": 1.0660589091223855e-05, | |
| "loss": 264.0925, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.3515948819607138, | |
| "grad_norm": 2897.310791015625, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 263.9691, | |
| "step": 15000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 5000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 0.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |