|
{ |
|
"best_metric": 0.42913728952407837, |
|
"best_model_checkpoint": "mikhail-panzo/fil_b32_le5_s8000/checkpoint-1500", |
|
"epoch": 33.333333333333336, |
|
"eval_steps": 500, |
|
"global_step": 1500, |
|
"is_hyper_param_search": false, |
|
"is_local_process_zero": true, |
|
"is_world_process_zero": true, |
|
"log_history": [ |
|
{ |
|
"epoch": 1.1111111111111112, |
|
"grad_norm": 3.579704523086548, |
|
"learning_rate": 2.4500000000000004e-07, |
|
"loss": 0.8282, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 2.2222222222222223, |
|
"grad_norm": 2.0918073654174805, |
|
"learning_rate": 4.95e-07, |
|
"loss": 0.7857, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 3.3333333333333335, |
|
"grad_norm": 1.283610224723816, |
|
"learning_rate": 7.450000000000001e-07, |
|
"loss": 0.7356, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 4.444444444444445, |
|
"grad_norm": 1.5381903648376465, |
|
"learning_rate": 9.950000000000002e-07, |
|
"loss": 0.7606, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 5.555555555555555, |
|
"grad_norm": 1.613647222518921, |
|
"learning_rate": 1.2450000000000002e-06, |
|
"loss": 0.7314, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 6.666666666666667, |
|
"grad_norm": 2.6762027740478516, |
|
"learning_rate": 1.495e-06, |
|
"loss": 0.6933, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 7.777777777777778, |
|
"grad_norm": 2.667527675628662, |
|
"learning_rate": 1.745e-06, |
|
"loss": 0.6863, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 8.88888888888889, |
|
"grad_norm": 1.9682470560073853, |
|
"learning_rate": 1.9950000000000004e-06, |
|
"loss": 0.6649, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 10.0, |
|
"grad_norm": 2.0042927265167236, |
|
"learning_rate": 2.245e-06, |
|
"loss": 0.6507, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 11.11111111111111, |
|
"grad_norm": 2.6399147510528564, |
|
"learning_rate": 2.4950000000000003e-06, |
|
"loss": 0.632, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 11.11111111111111, |
|
"eval_loss": 0.5323343276977539, |
|
"eval_runtime": 7.6175, |
|
"eval_samples_per_second": 20.873, |
|
"eval_steps_per_second": 2.626, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 12.222222222222221, |
|
"grad_norm": 2.593517780303955, |
|
"learning_rate": 2.7450000000000004e-06, |
|
"loss": 0.5956, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 13.333333333333334, |
|
"grad_norm": 5.524081707000732, |
|
"learning_rate": 2.99e-06, |
|
"loss": 0.577, |
|
"step": 600 |
|
}, |
|
{ |
|
"epoch": 14.444444444444445, |
|
"grad_norm": 2.2165024280548096, |
|
"learning_rate": 3.2400000000000003e-06, |
|
"loss": 0.5614, |
|
"step": 650 |
|
}, |
|
{ |
|
"epoch": 15.555555555555555, |
|
"grad_norm": 2.4440901279449463, |
|
"learning_rate": 3.49e-06, |
|
"loss": 0.5654, |
|
"step": 700 |
|
}, |
|
{ |
|
"epoch": 16.666666666666668, |
|
"grad_norm": 1.8401445150375366, |
|
"learning_rate": 3.74e-06, |
|
"loss": 0.5329, |
|
"step": 750 |
|
}, |
|
{ |
|
"epoch": 17.77777777777778, |
|
"grad_norm": 2.101787567138672, |
|
"learning_rate": 3.990000000000001e-06, |
|
"loss": 0.5293, |
|
"step": 800 |
|
}, |
|
{ |
|
"epoch": 18.88888888888889, |
|
"grad_norm": 2.1338887214660645, |
|
"learning_rate": 4.24e-06, |
|
"loss": 0.5345, |
|
"step": 850 |
|
}, |
|
{ |
|
"epoch": 20.0, |
|
"grad_norm": 1.8499983549118042, |
|
"learning_rate": 4.49e-06, |
|
"loss": 0.5148, |
|
"step": 900 |
|
}, |
|
{ |
|
"epoch": 21.11111111111111, |
|
"grad_norm": 1.3982588052749634, |
|
"learning_rate": 4.74e-06, |
|
"loss": 0.5085, |
|
"step": 950 |
|
}, |
|
{ |
|
"epoch": 22.22222222222222, |
|
"grad_norm": 1.7611488103866577, |
|
"learning_rate": 4.9900000000000005e-06, |
|
"loss": 0.519, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 22.22222222222222, |
|
"eval_loss": 0.4493897557258606, |
|
"eval_runtime": 7.4167, |
|
"eval_samples_per_second": 21.438, |
|
"eval_steps_per_second": 2.697, |
|
"step": 1000 |
|
}, |
|
{ |
|
"epoch": 23.333333333333332, |
|
"grad_norm": 2.625955581665039, |
|
"learning_rate": 5.240000000000001e-06, |
|
"loss": 0.5194, |
|
"step": 1050 |
|
}, |
|
{ |
|
"epoch": 24.444444444444443, |
|
"grad_norm": 1.75005304813385, |
|
"learning_rate": 5.490000000000001e-06, |
|
"loss": 0.5152, |
|
"step": 1100 |
|
}, |
|
{ |
|
"epoch": 25.555555555555557, |
|
"grad_norm": 2.279965877532959, |
|
"learning_rate": 5.74e-06, |
|
"loss": 0.5129, |
|
"step": 1150 |
|
}, |
|
{ |
|
"epoch": 26.666666666666668, |
|
"grad_norm": 1.5506455898284912, |
|
"learning_rate": 5.99e-06, |
|
"loss": 0.4941, |
|
"step": 1200 |
|
}, |
|
{ |
|
"epoch": 27.77777777777778, |
|
"grad_norm": 1.830841064453125, |
|
"learning_rate": 6.24e-06, |
|
"loss": 0.4899, |
|
"step": 1250 |
|
}, |
|
{ |
|
"epoch": 28.88888888888889, |
|
"grad_norm": 2.0044639110565186, |
|
"learning_rate": 6.4900000000000005e-06, |
|
"loss": 0.496, |
|
"step": 1300 |
|
}, |
|
{ |
|
"epoch": 30.0, |
|
"grad_norm": 2.272815465927124, |
|
"learning_rate": 6.740000000000001e-06, |
|
"loss": 0.4906, |
|
"step": 1350 |
|
}, |
|
{ |
|
"epoch": 31.11111111111111, |
|
"grad_norm": 1.460723638534546, |
|
"learning_rate": 6.99e-06, |
|
"loss": 0.4741, |
|
"step": 1400 |
|
}, |
|
{ |
|
"epoch": 32.22222222222222, |
|
"grad_norm": 1.7440383434295654, |
|
"learning_rate": 7.24e-06, |
|
"loss": 0.4821, |
|
"step": 1450 |
|
}, |
|
{ |
|
"epoch": 33.333333333333336, |
|
"grad_norm": 1.4221312999725342, |
|
"learning_rate": 7.49e-06, |
|
"loss": 0.4816, |
|
"step": 1500 |
|
}, |
|
{ |
|
"epoch": 33.333333333333336, |
|
"eval_loss": 0.42913728952407837, |
|
"eval_runtime": 7.4808, |
|
"eval_samples_per_second": 21.254, |
|
"eval_steps_per_second": 2.674, |
|
"step": 1500 |
|
} |
|
], |
|
"logging_steps": 50, |
|
"max_steps": 8000, |
|
"num_input_tokens_seen": 0, |
|
"num_train_epochs": 178, |
|
"save_steps": 500, |
|
"stateful_callbacks": { |
|
"TrainerControl": { |
|
"args": { |
|
"should_epoch_stop": false, |
|
"should_evaluate": false, |
|
"should_log": false, |
|
"should_save": true, |
|
"should_training_stop": false |
|
}, |
|
"attributes": {} |
|
} |
|
}, |
|
"total_flos": 1.0948113803068992e+16, |
|
"train_batch_size": 32, |
|
"trial_name": null, |
|
"trial_params": null |
|
} |
|
|