| { | |
| "best_metric": 43.53, | |
| "best_model_checkpoint": "opus_big_lsp_adapt_wce_precision_3_ubweight_1.5/checkpoint-80000", | |
| "epoch": 1.9716574245224892, | |
| "global_step": 80000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.9753635243376465e-05, | |
| "loss": 0.1796, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_bleu": 42.245, | |
| "eval_gen_len": 33.5858, | |
| "eval_loss": 0.09831776469945908, | |
| "eval_runtime": 132.7068, | |
| "eval_samples_per_second": 7.859, | |
| "eval_steps_per_second": 0.497, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.9507208872458414e-05, | |
| "loss": 0.1487, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "eval_bleu": 42.2116, | |
| "eval_gen_len": 34.3643, | |
| "eval_loss": 0.09730643033981323, | |
| "eval_runtime": 116.1316, | |
| "eval_samples_per_second": 8.981, | |
| "eval_steps_per_second": 0.568, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 1.9260844115834877e-05, | |
| "loss": 0.1446, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "eval_bleu": 42.0306, | |
| "eval_gen_len": 33.4008, | |
| "eval_loss": 0.09792491048574448, | |
| "eval_runtime": 139.3867, | |
| "eval_samples_per_second": 7.483, | |
| "eval_steps_per_second": 0.474, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.901447935921134e-05, | |
| "loss": 0.1406, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_bleu": 42.7137, | |
| "eval_gen_len": 34.441, | |
| "eval_loss": 0.0964345782995224, | |
| "eval_runtime": 121.1927, | |
| "eval_samples_per_second": 8.606, | |
| "eval_steps_per_second": 0.545, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 1.8768114602587803e-05, | |
| "loss": 0.1391, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_bleu": 42.8852, | |
| "eval_gen_len": 33.8102, | |
| "eval_loss": 0.09575977176427841, | |
| "eval_runtime": 107.5204, | |
| "eval_samples_per_second": 9.7, | |
| "eval_steps_per_second": 0.614, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 1.8521749845964266e-05, | |
| "loss": 0.1239, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "eval_bleu": 42.9831, | |
| "eval_gen_len": 34.3087, | |
| "eval_loss": 0.09662675112485886, | |
| "eval_runtime": 114.4632, | |
| "eval_samples_per_second": 9.112, | |
| "eval_steps_per_second": 0.577, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 1.827538508934073e-05, | |
| "loss": 0.1242, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "eval_bleu": 42.7598, | |
| "eval_gen_len": 34.2502, | |
| "eval_loss": 0.09706231206655502, | |
| "eval_runtime": 120.5178, | |
| "eval_samples_per_second": 8.654, | |
| "eval_steps_per_second": 0.548, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 1.8029020332717192e-05, | |
| "loss": 0.1248, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_bleu": 42.9725, | |
| "eval_gen_len": 34.6242, | |
| "eval_loss": 0.09721297025680542, | |
| "eval_runtime": 217.1644, | |
| "eval_samples_per_second": 4.803, | |
| "eval_steps_per_second": 0.304, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 1.7782655576093655e-05, | |
| "loss": 0.1243, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_bleu": 42.7137, | |
| "eval_gen_len": 34.441, | |
| "eval_loss": 0.09667866677045822, | |
| "eval_runtime": 206.9894, | |
| "eval_samples_per_second": 5.039, | |
| "eval_steps_per_second": 0.319, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "learning_rate": 1.753629081947012e-05, | |
| "loss": 0.1241, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_bleu": 43.53, | |
| "eval_gen_len": 33.9923, | |
| "eval_loss": 0.0961531400680542, | |
| "eval_runtime": 111.5933, | |
| "eval_samples_per_second": 9.346, | |
| "eval_steps_per_second": 0.591, | |
| "step": 80000 | |
| } | |
| ], | |
| "max_steps": 649200, | |
| "num_train_epochs": 16, | |
| "total_flos": 2.5844916958632346e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |