| { | |
| "best_metric": 0.7868753671646118, | |
| "best_model_checkpoint": "./Zeroshot/01-12-23-mistralai-Mistral-7B-v0.1_multilang-dataset-3.0.3-portuguese-2_epochs-10_batch_3/checkpoints/checkpoint-27092", | |
| "epoch": 8.999852354938728, | |
| "eval_steps": 500, | |
| "global_step": 30478, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 5.835794447725931e-05, | |
| "loss": 1.4468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 0.00011742468989958655, | |
| "loss": 0.9754, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.00017649143532191377, | |
| "loss": 0.9429, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.000235558180744241, | |
| "loss": 0.9147, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 0.00029462492616656825, | |
| "loss": 0.9067, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.00035369167158889544, | |
| "loss": 0.8978, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.8805813789367676, | |
| "eval_runtime": 88.2945, | |
| "eval_samples_per_second": 17.057, | |
| "eval_steps_per_second": 2.141, | |
| "step": 3386 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 0.00039998760393503537, | |
| "loss": 0.895, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.0003996072594095129, | |
| "loss": 0.8687, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "learning_rate": 0.00039869668890858337, | |
| "loss": 0.8884, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 0.00039725831122269285, | |
| "loss": 0.8715, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.00039529594718087214, | |
| "loss": 0.8645, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0003928148095012922, | |
| "loss": 0.8666, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.0003898214889444803, | |
| "loss": 0.8719, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.8552550673484802, | |
| "eval_runtime": 88.3232, | |
| "eval_samples_per_second": 17.051, | |
| "eval_steps_per_second": 2.14, | |
| "step": 6773 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "learning_rate": 0.00038632393680597854, | |
| "loss": 0.8438, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 0.0003823314437949511, | |
| "loss": 0.8308, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 0.00037785461535484375, | |
| "loss": 0.8259, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 0.0003729053434916558, | |
| "loss": 0.8324, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.66, | |
| "learning_rate": 0.0003674967751846552, | |
| "loss": 0.8413, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 0.0003616554183563445, | |
| "loss": 0.8322, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 0.00035537338261496887, | |
| "loss": 0.8368, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.8443310260772705, | |
| "eval_runtime": 88.3102, | |
| "eval_samples_per_second": 17.054, | |
| "eval_steps_per_second": 2.14, | |
| "step": 10159 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 0.0003486786213865893, | |
| "loss": 0.8088, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 3.25, | |
| "learning_rate": 0.0003415889182744321, | |
| "loss": 0.8003, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 0.0003341231059840768, | |
| "loss": 0.805, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 3.54, | |
| "learning_rate": 0.0003263010162972709, | |
| "loss": 0.8061, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 0.00031814342739185336, | |
| "loss": 0.8008, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "learning_rate": 0.000309672008647721, | |
| "loss": 0.8029, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00030090926308545536, | |
| "loss": 0.8056, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8322489857673645, | |
| "eval_runtime": 88.3294, | |
| "eval_samples_per_second": 17.05, | |
| "eval_steps_per_second": 2.14, | |
| "step": 13546 | |
| }, | |
| { | |
| "epoch": 4.13, | |
| "learning_rate": 0.00029187846759051, | |
| "loss": 0.7649, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 4.28, | |
| "learning_rate": 0.00028260361108174584, | |
| "loss": 0.7674, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 0.0002731093307885585, | |
| "loss": 0.7635, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 4.58, | |
| "learning_rate": 0.0002634208468058692, | |
| "loss": 0.7759, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 4.72, | |
| "learning_rate": 0.000253563895100822, | |
| "loss": 0.7669, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 4.87, | |
| "learning_rate": 0.0002435847820221107, | |
| "loss": 0.7708, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.8134331107139587, | |
| "eval_runtime": 88.3002, | |
| "eval_samples_per_second": 17.055, | |
| "eval_steps_per_second": 2.14, | |
| "step": 16932 | |
| }, | |
| { | |
| "epoch": 5.02, | |
| "learning_rate": 0.00023347002797691627, | |
| "loss": 0.7681, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 5.17, | |
| "learning_rate": 0.00022326636595170415, | |
| "loss": 0.7237, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 5.32, | |
| "learning_rate": 0.0002130214749113661, | |
| "loss": 0.7299, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 0.0002027215162553563, | |
| "loss": 0.7307, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 0.0001924143283101145, | |
| "loss": 0.7251, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 5.76, | |
| "learning_rate": 0.00018212729053467496, | |
| "loss": 0.7328, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.00017188772886224626, | |
| "loss": 0.731, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.8020778298377991, | |
| "eval_runtime": 88.3249, | |
| "eval_samples_per_second": 17.051, | |
| "eval_steps_per_second": 2.14, | |
| "step": 20319 | |
| }, | |
| { | |
| "epoch": 6.05, | |
| "learning_rate": 0.00016172284311307314, | |
| "loss": 0.7118, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "learning_rate": 0.00015167964186544786, | |
| "loss": 0.691, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 6.35, | |
| "learning_rate": 0.00014174455895749337, | |
| "loss": 0.6873, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 0.00013196422266617455, | |
| "loss": 0.6892, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 0.00012236461295016225, | |
| "loss": 0.6864, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 6.79, | |
| "learning_rate": 0.00011298979396555838, | |
| "loss": 0.6848, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 6.94, | |
| "learning_rate": 0.00010382710238831153, | |
| "loss": 0.6828, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.7894787788391113, | |
| "eval_runtime": 88.3291, | |
| "eval_samples_per_second": 17.05, | |
| "eval_steps_per_second": 2.14, | |
| "step": 23705 | |
| }, | |
| { | |
| "epoch": 7.09, | |
| "learning_rate": 9.491987931754444e-05, | |
| "loss": 0.6629, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 7.23, | |
| "learning_rate": 8.629178542098462e-05, | |
| "loss": 0.652, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 7.38, | |
| "learning_rate": 7.796573990265166e-05, | |
| "loss": 0.6479, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 7.53, | |
| "learning_rate": 6.996385962151814e-05, | |
| "loss": 0.65, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 7.68, | |
| "learning_rate": 6.230740034147598e-05, | |
| "loss": 0.6455, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 5.503090360176009e-05, | |
| "loss": 0.6516, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 7.97, | |
| "learning_rate": 4.8124540451398006e-05, | |
| "loss": 0.6453, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7868753671646118, | |
| "eval_runtime": 88.3103, | |
| "eval_samples_per_second": 17.054, | |
| "eval_steps_per_second": 2.14, | |
| "step": 27092 | |
| }, | |
| { | |
| "epoch": 8.12, | |
| "learning_rate": 4.162161109356324e-05, | |
| "loss": 0.6222, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 8.27, | |
| "learning_rate": 3.553938955963088e-05, | |
| "loss": 0.6164, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 8.42, | |
| "learning_rate": 2.9894032335367272e-05, | |
| "loss": 0.6214, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 8.56, | |
| "learning_rate": 2.470053544371056e-05, | |
| "loss": 0.6215, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 1.9981677010252242e-05, | |
| "loss": 0.6181, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 1.573108289234544e-05, | |
| "loss": 0.6179, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.7876124382019043, | |
| "eval_runtime": 88.313, | |
| "eval_samples_per_second": 17.053, | |
| "eval_steps_per_second": 2.14, | |
| "step": 30478 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 33860, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.530218263203152e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |