| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0036865510011290063, | |
| "eval_steps": 5, | |
| "global_step": 20, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0001843275500564503, | |
| "grad_norm": 0.10575726628303528, | |
| "learning_rate": 1e-05, | |
| "loss": 6.9357, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0001843275500564503, | |
| "eval_loss": 6.940097808837891, | |
| "eval_runtime": 41.232, | |
| "eval_samples_per_second": 221.6, | |
| "eval_steps_per_second": 110.812, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0003686551001129006, | |
| "grad_norm": 0.09431350976228714, | |
| "learning_rate": 2e-05, | |
| "loss": 6.9415, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0005529826501693509, | |
| "grad_norm": 0.09502771496772766, | |
| "learning_rate": 3e-05, | |
| "loss": 6.943, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0007373102002258012, | |
| "grad_norm": 0.09998767077922821, | |
| "learning_rate": 4e-05, | |
| "loss": 6.9373, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0009216377502822516, | |
| "grad_norm": 0.10265155881643295, | |
| "learning_rate": 5e-05, | |
| "loss": 6.9364, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0009216377502822516, | |
| "eval_loss": 6.939851760864258, | |
| "eval_runtime": 16.2124, | |
| "eval_samples_per_second": 563.582, | |
| "eval_steps_per_second": 281.822, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0011059653003387018, | |
| "grad_norm": 0.10776941478252411, | |
| "learning_rate": 6e-05, | |
| "loss": 6.9381, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0012902928503951523, | |
| "grad_norm": 0.1045922115445137, | |
| "learning_rate": 7e-05, | |
| "loss": 6.9417, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0014746204004516025, | |
| "grad_norm": 0.09844892472028732, | |
| "learning_rate": 8e-05, | |
| "loss": 6.9342, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.0016589479505080527, | |
| "grad_norm": 0.10482911020517349, | |
| "learning_rate": 9e-05, | |
| "loss": 6.9423, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0018432755005645032, | |
| "grad_norm": 0.10163327306509018, | |
| "learning_rate": 0.0001, | |
| "loss": 6.9399, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0018432755005645032, | |
| "eval_loss": 6.9390177726745605, | |
| "eval_runtime": 12.1573, | |
| "eval_samples_per_second": 751.565, | |
| "eval_steps_per_second": 375.823, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0020276030506209534, | |
| "grad_norm": 0.10278677940368652, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 6.9418, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.0022119306006774036, | |
| "grad_norm": 0.105440154671669, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 6.9387, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.002396258150733854, | |
| "grad_norm": 0.09692682325839996, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 6.9364, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.0025805857007903045, | |
| "grad_norm": 0.10328719764947891, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 6.9379, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0027649132508467547, | |
| "grad_norm": 0.09780015796422958, | |
| "learning_rate": 5e-05, | |
| "loss": 6.934, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0027649132508467547, | |
| "eval_loss": 6.937956809997559, | |
| "eval_runtime": 12.06, | |
| "eval_samples_per_second": 757.629, | |
| "eval_steps_per_second": 378.856, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.002949240800903205, | |
| "grad_norm": 0.10987330228090286, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 6.9401, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.003133568350959655, | |
| "grad_norm": 0.11077705770730972, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 6.9358, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0033178959010161054, | |
| "grad_norm": 0.10548257827758789, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 6.937, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.003502223451072556, | |
| "grad_norm": 0.1043916642665863, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 6.9403, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0036865510011290063, | |
| "grad_norm": 0.10964982956647873, | |
| "learning_rate": 0.0, | |
| "loss": 6.9381, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0036865510011290063, | |
| "eval_loss": 6.937661170959473, | |
| "eval_runtime": 12.1995, | |
| "eval_samples_per_second": 748.968, | |
| "eval_steps_per_second": 374.525, | |
| "step": 20 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 20, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 5, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 105796012032.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |