|
{ |
|
"best_metric": null, |
|
"best_model_checkpoint": null, |
|
"epoch": 4.0, |
|
"eval_steps": 25, |
|
"global_step": 36, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 0.1111111111111111, |
|
"grad_norm": 103557.7421875, |
|
"learning_rate": 0.00019444444444444446, |
|
"loss": 9.6138, |
|
"step": 1 |
|
}, |
|
{ |
|
"epoch": 0.2222222222222222, |
|
"grad_norm": 91293.6640625, |
|
"learning_rate": 0.00018888888888888888, |
|
"loss": 9.4432, |
|
"step": 2 |
|
}, |
|
{ |
|
"epoch": 0.3333333333333333, |
|
"grad_norm": 112938.5078125, |
|
"learning_rate": 0.00018333333333333334, |
|
"loss": 9.2664, |
|
"step": 3 |
|
}, |
|
{ |
|
"epoch": 0.4444444444444444, |
|
"grad_norm": 125412.2890625, |
|
"learning_rate": 0.00017777777777777779, |
|
"loss": 9.1266, |
|
"step": 4 |
|
}, |
|
{ |
|
"epoch": 0.5555555555555556, |
|
"grad_norm": 153359.796875, |
|
"learning_rate": 0.00017222222222222224, |
|
"loss": 8.821, |
|
"step": 5 |
|
}, |
|
{ |
|
"epoch": 0.6666666666666666, |
|
"grad_norm": 369198.6875, |
|
"learning_rate": 0.0001666666666666667, |
|
"loss": 8.6783, |
|
"step": 6 |
|
}, |
|
{ |
|
"epoch": 0.7777777777777778, |
|
"grad_norm": 220549.6875, |
|
"learning_rate": 0.0001611111111111111, |
|
"loss": 8.1081, |
|
"step": 7 |
|
}, |
|
{ |
|
"epoch": 0.8888888888888888, |
|
"grad_norm": 167152.3125, |
|
"learning_rate": 0.00015555555555555556, |
|
"loss": 7.9356, |
|
"step": 8 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 218152.984375, |
|
"learning_rate": 0.00015000000000000001, |
|
"loss": 8.0719, |
|
"step": 9 |
|
}, |
|
{ |
|
"epoch": 1.1111111111111112, |
|
"grad_norm": 109639.3984375, |
|
"learning_rate": 0.00014444444444444444, |
|
"loss": 7.6857, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 1.2222222222222223, |
|
"grad_norm": 154048.453125, |
|
"learning_rate": 0.0001388888888888889, |
|
"loss": 7.7333, |
|
"step": 11 |
|
}, |
|
{ |
|
"epoch": 1.3333333333333333, |
|
"grad_norm": 77620.4609375, |
|
"learning_rate": 0.00013333333333333334, |
|
"loss": 7.5804, |
|
"step": 12 |
|
}, |
|
{ |
|
"epoch": 1.4444444444444444, |
|
"grad_norm": 121529.203125, |
|
"learning_rate": 0.00012777777777777776, |
|
"loss": 7.3028, |
|
"step": 13 |
|
}, |
|
{ |
|
"epoch": 1.5555555555555556, |
|
"grad_norm": 95948.5546875, |
|
"learning_rate": 0.00012222222222222224, |
|
"loss": 7.3862, |
|
"step": 14 |
|
}, |
|
{ |
|
"epoch": 1.6666666666666665, |
|
"grad_norm": 58191.4765625, |
|
"learning_rate": 0.00011666666666666668, |
|
"loss": 7.3828, |
|
"step": 15 |
|
}, |
|
{ |
|
"epoch": 1.7777777777777777, |
|
"grad_norm": 117447.7890625, |
|
"learning_rate": 0.00011111111111111112, |
|
"loss": 7.4322, |
|
"step": 16 |
|
}, |
|
{ |
|
"epoch": 1.8888888888888888, |
|
"grad_norm": 89299.1953125, |
|
"learning_rate": 0.00010555555555555557, |
|
"loss": 7.3808, |
|
"step": 17 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 129674.390625, |
|
"learning_rate": 0.0001, |
|
"loss": 7.1292, |
|
"step": 18 |
|
}, |
|
{ |
|
"epoch": 2.111111111111111, |
|
"grad_norm": 79440.0546875, |
|
"learning_rate": 9.444444444444444e-05, |
|
"loss": 7.2321, |
|
"step": 19 |
|
}, |
|
{ |
|
"epoch": 2.2222222222222223, |
|
"grad_norm": 118613.7109375, |
|
"learning_rate": 8.888888888888889e-05, |
|
"loss": 7.368, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 2.3333333333333335, |
|
"grad_norm": 383910.0, |
|
"learning_rate": 8.333333333333334e-05, |
|
"loss": 7.2749, |
|
"step": 21 |
|
}, |
|
{ |
|
"epoch": 2.4444444444444446, |
|
"grad_norm": 87922.9140625, |
|
"learning_rate": 7.777777777777778e-05, |
|
"loss": 7.3671, |
|
"step": 22 |
|
}, |
|
{ |
|
"epoch": 2.5555555555555554, |
|
"grad_norm": 142905.296875, |
|
"learning_rate": 7.222222222222222e-05, |
|
"loss": 7.4515, |
|
"step": 23 |
|
}, |
|
{ |
|
"epoch": 2.6666666666666665, |
|
"grad_norm": 118390.03125, |
|
"learning_rate": 6.666666666666667e-05, |
|
"loss": 7.2825, |
|
"step": 24 |
|
}, |
|
{ |
|
"epoch": 2.7777777777777777, |
|
"grad_norm": 102163.3125, |
|
"learning_rate": 6.111111111111112e-05, |
|
"loss": 7.1679, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 2.7777777777777777, |
|
"eval_clap": -0.0018186935922130942, |
|
"eval_loss": 6.224447727203369, |
|
"eval_runtime": 50.224, |
|
"eval_samples_per_second": 0.1, |
|
"eval_steps_per_second": 0.04, |
|
"step": 25 |
|
}, |
|
{ |
|
"epoch": 2.888888888888889, |
|
"grad_norm": 50973.87109375, |
|
"learning_rate": 5.555555555555556e-05, |
|
"loss": 7.2655, |
|
"step": 26 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 105052.3046875, |
|
"learning_rate": 5e-05, |
|
"loss": 7.2949, |
|
"step": 27 |
|
}, |
|
{ |
|
"epoch": 3.111111111111111, |
|
"grad_norm": 93693.15625, |
|
"learning_rate": 4.4444444444444447e-05, |
|
"loss": 7.3318, |
|
"step": 28 |
|
}, |
|
{ |
|
"epoch": 3.2222222222222223, |
|
"grad_norm": 50967.5078125, |
|
"learning_rate": 3.888888888888889e-05, |
|
"loss": 7.219, |
|
"step": 29 |
|
}, |
|
{ |
|
"epoch": 3.3333333333333335, |
|
"grad_norm": 119510.9921875, |
|
"learning_rate": 3.3333333333333335e-05, |
|
"loss": 7.1688, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 3.4444444444444446, |
|
"grad_norm": 85723.8359375, |
|
"learning_rate": 2.777777777777778e-05, |
|
"loss": 7.1438, |
|
"step": 31 |
|
}, |
|
{ |
|
"epoch": 3.5555555555555554, |
|
"grad_norm": 79061.3671875, |
|
"learning_rate": 2.2222222222222223e-05, |
|
"loss": 7.1922, |
|
"step": 32 |
|
}, |
|
{ |
|
"epoch": 3.6666666666666665, |
|
"grad_norm": 71475.859375, |
|
"learning_rate": 1.6666666666666667e-05, |
|
"loss": 7.3682, |
|
"step": 33 |
|
}, |
|
{ |
|
"epoch": 3.7777777777777777, |
|
"grad_norm": 90030.9921875, |
|
"learning_rate": 1.1111111111111112e-05, |
|
"loss": 7.3219, |
|
"step": 34 |
|
}, |
|
{ |
|
"epoch": 3.888888888888889, |
|
"grad_norm": 62880.04296875, |
|
"learning_rate": 5.555555555555556e-06, |
|
"loss": 7.2549, |
|
"step": 35 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"grad_norm": 85328.8671875, |
|
"learning_rate": 0.0, |
|
"loss": 7.2549, |
|
"step": 36 |
|
}, |
|
{ |
|
"epoch": 4.0, |
|
"step": 36, |
|
"total_flos": 3658505195664.0, |
|
"train_loss": 7.6954962147606745, |
|
"train_runtime": 1412.8506, |
|
"train_samples_per_second": 0.093, |
|
"train_steps_per_second": 0.025 |
|
} |
|
], |
|
"logging_steps": 1.0, |
|
"max_steps": 36, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 4, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": false, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 3658505195664.0, |
|
"train_batch_size": 4, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|