| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10150891632373114, | |
| "eval_steps": 500, | |
| "global_step": 37, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0027434842249657062, | |
| "grad_norm": 1.902702808380127, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 2.0959, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0054869684499314125, | |
| "grad_norm": 1.8936893939971924, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 2.099, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.00823045267489712, | |
| "grad_norm": 1.9235605001449585, | |
| "learning_rate": 6e-06, | |
| "loss": 2.1045, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.010973936899862825, | |
| "grad_norm": 1.7873843908309937, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.0144, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.013717421124828532, | |
| "grad_norm": 1.8321126699447632, | |
| "learning_rate": 1e-05, | |
| "loss": 2.0872, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.01646090534979424, | |
| "grad_norm": 2.0333194732666016, | |
| "learning_rate": 1.2e-05, | |
| "loss": 2.1146, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.019204389574759947, | |
| "grad_norm": 1.763102650642395, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 2.036, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.02194787379972565, | |
| "grad_norm": 1.5287824869155884, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.984, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.024691358024691357, | |
| "grad_norm": 1.2035481929779053, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.928, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.027434842249657063, | |
| "grad_norm": 1.0602883100509644, | |
| "learning_rate": 2e-05, | |
| "loss": 1.9322, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.03017832647462277, | |
| "grad_norm": 0.9456723928451538, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.9257, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.03292181069958848, | |
| "grad_norm": 0.7646191120147705, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.8057, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.03566529492455418, | |
| "grad_norm": 0.6718866229057312, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.8535, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.038408779149519894, | |
| "grad_norm": 0.5136308073997498, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.8533, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0411522633744856, | |
| "grad_norm": 0.4066430628299713, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7867, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0438957475994513, | |
| "grad_norm": 0.3591379225254059, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.8729, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.04663923182441701, | |
| "grad_norm": 0.3689568340778351, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.8186, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.04938271604938271, | |
| "grad_norm": 0.3991197645664215, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.8485, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.05212620027434842, | |
| "grad_norm": 0.3603827655315399, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.8081, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.05486968449931413, | |
| "grad_norm": 0.3805089592933655, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7553, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.05761316872427984, | |
| "grad_norm": 0.42597419023513794, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.8115, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.06035665294924554, | |
| "grad_norm": 0.37801623344421387, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.7161, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.06310013717421124, | |
| "grad_norm": 0.35600143671035767, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.7315, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.06584362139917696, | |
| "grad_norm": 0.4111214280128479, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.7091, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.06858710562414266, | |
| "grad_norm": 0.4117395579814911, | |
| "learning_rate": 5e-05, | |
| "loss": 1.7026, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.07133058984910837, | |
| "grad_norm": 0.4069993197917938, | |
| "learning_rate": 5.2000000000000004e-05, | |
| "loss": 1.6733, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.07407407407407407, | |
| "grad_norm": 0.4196024239063263, | |
| "learning_rate": 5.4000000000000005e-05, | |
| "loss": 1.6999, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.07681755829903979, | |
| "grad_norm": 0.45915013551712036, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.6613, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.07956104252400549, | |
| "grad_norm": 0.44673749804496765, | |
| "learning_rate": 5.8e-05, | |
| "loss": 1.7039, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0823045267489712, | |
| "grad_norm": 0.5381506085395813, | |
| "learning_rate": 6e-05, | |
| "loss": 1.5768, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0850480109739369, | |
| "grad_norm": 0.4311385750770569, | |
| "learning_rate": 6.2e-05, | |
| "loss": 1.5598, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0877914951989026, | |
| "grad_norm": 0.38209667801856995, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.511, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.09053497942386832, | |
| "grad_norm": 0.35374918580055237, | |
| "learning_rate": 6.6e-05, | |
| "loss": 1.4692, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.09327846364883402, | |
| "grad_norm": 0.4020269215106964, | |
| "learning_rate": 6.800000000000001e-05, | |
| "loss": 1.5276, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.09602194787379972, | |
| "grad_norm": 0.46055856347084045, | |
| "learning_rate": 7e-05, | |
| "loss": 1.4634, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.09876543209876543, | |
| "grad_norm": 0.4301123023033142, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.4151, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.10150891632373114, | |
| "grad_norm": 0.34214499592781067, | |
| "learning_rate": 7.4e-05, | |
| "loss": 1.4451, | |
| "step": 37 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 364, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 1, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.56958749169705e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |