| { | |
| "best_metric": 0.4933945834636688, | |
| "best_model_checkpoint": "mikhail_panzo/zlm_b64_le5_s12000/checkpoint-1000", | |
| "epoch": 0.837696335078534, | |
| "eval_steps": 500, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.041884816753926704, | |
| "grad_norm": 12.821951866149902, | |
| "learning_rate": 2.4000000000000003e-07, | |
| "loss": 1.1152, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08376963350785341, | |
| "grad_norm": 6.186221122741699, | |
| "learning_rate": 4.900000000000001e-07, | |
| "loss": 1.0584, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1256544502617801, | |
| "grad_norm": 13.08674430847168, | |
| "learning_rate": 7.4e-07, | |
| "loss": 0.9996, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.16753926701570682, | |
| "grad_norm": 4.10632848739624, | |
| "learning_rate": 9.9e-07, | |
| "loss": 0.9355, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2094240837696335, | |
| "grad_norm": 3.2428064346313477, | |
| "learning_rate": 1.2400000000000002e-06, | |
| "loss": 0.852, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.2513089005235602, | |
| "grad_norm": 2.0861401557922363, | |
| "learning_rate": 1.4900000000000001e-06, | |
| "loss": 0.8241, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.2931937172774869, | |
| "grad_norm": 8.602317810058594, | |
| "learning_rate": 1.74e-06, | |
| "loss": 0.7629, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.33507853403141363, | |
| "grad_norm": 2.7594990730285645, | |
| "learning_rate": 1.9900000000000004e-06, | |
| "loss": 0.7418, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.3769633507853403, | |
| "grad_norm": 2.7047348022460938, | |
| "learning_rate": 2.24e-06, | |
| "loss": 0.7166, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "grad_norm": 2.0666158199310303, | |
| "learning_rate": 2.4900000000000003e-06, | |
| "loss": 0.6876, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.418848167539267, | |
| "eval_loss": 0.5937426090240479, | |
| "eval_runtime": 263.0955, | |
| "eval_samples_per_second": 32.266, | |
| "eval_steps_per_second": 4.037, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4607329842931937, | |
| "grad_norm": 2.955193519592285, | |
| "learning_rate": 2.7400000000000004e-06, | |
| "loss": 0.673, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.5026178010471204, | |
| "grad_norm": 4.856322765350342, | |
| "learning_rate": 2.99e-06, | |
| "loss": 0.6539, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5445026178010471, | |
| "grad_norm": 7.631103992462158, | |
| "learning_rate": 3.2400000000000003e-06, | |
| "loss": 0.6588, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.5863874345549738, | |
| "grad_norm": 2.715771436691284, | |
| "learning_rate": 3.49e-06, | |
| "loss": 0.624, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.6282722513089005, | |
| "grad_norm": 2.7684273719787598, | |
| "learning_rate": 3.74e-06, | |
| "loss": 0.6116, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.6701570680628273, | |
| "grad_norm": 2.219599723815918, | |
| "learning_rate": 3.990000000000001e-06, | |
| "loss": 0.6097, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.7120418848167539, | |
| "grad_norm": 3.749188184738159, | |
| "learning_rate": 4.24e-06, | |
| "loss": 0.614, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.7539267015706806, | |
| "grad_norm": 2.120619773864746, | |
| "learning_rate": 4.49e-06, | |
| "loss": 0.5814, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.7958115183246073, | |
| "grad_norm": 2.7725327014923096, | |
| "learning_rate": 4.74e-06, | |
| "loss": 0.5895, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.837696335078534, | |
| "grad_norm": 9.851961135864258, | |
| "learning_rate": 4.9900000000000005e-06, | |
| "loss": 0.5623, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.837696335078534, | |
| "eval_loss": 0.4933945834636688, | |
| "eval_runtime": 258.739, | |
| "eval_samples_per_second": 32.809, | |
| "eval_steps_per_second": 4.105, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 12000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 11, | |
| "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": 9000282308548608.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |