| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.6, | |
| "eval_steps": 10, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 9.971570014953613, | |
| "learning_rate": 0.00019, | |
| "loss": 0.3706, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 4.22176456451416, | |
| "learning_rate": 0.00018, | |
| "loss": 0.3124, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 0.3505728542804718, | |
| "eval_runtime": 64.0226, | |
| "eval_samples_per_second": 3.124, | |
| "eval_steps_per_second": 1.562, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 3.4927077293395996, | |
| "learning_rate": 0.00017, | |
| "loss": 0.2459, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 5.083699703216553, | |
| "learning_rate": 0.00016, | |
| "loss": 0.2418, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "eval_loss": 0.3516603112220764, | |
| "eval_runtime": 63.9821, | |
| "eval_samples_per_second": 3.126, | |
| "eval_steps_per_second": 1.563, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.345269680023193, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.2474, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 3.5892534255981445, | |
| "learning_rate": 0.00014, | |
| "loss": 0.2916, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 0.34584036469459534, | |
| "eval_runtime": 64.416, | |
| "eval_samples_per_second": 3.105, | |
| "eval_steps_per_second": 1.552, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 3.6377668380737305, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.2859, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 3.6360130310058594, | |
| "learning_rate": 0.00012, | |
| "loss": 0.2943, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "eval_loss": 0.33658257126808167, | |
| "eval_runtime": 64.2616, | |
| "eval_samples_per_second": 3.112, | |
| "eval_steps_per_second": 1.556, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 4.8437323570251465, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 0.2664, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.5033814907073975, | |
| "learning_rate": 0.0001, | |
| "loss": 0.2793, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 0.34766554832458496, | |
| "eval_runtime": 64.2152, | |
| "eval_samples_per_second": 3.115, | |
| "eval_steps_per_second": 1.557, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 3.4198801517486572, | |
| "learning_rate": 9e-05, | |
| "loss": 0.2661, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 3.3779234886169434, | |
| "learning_rate": 8e-05, | |
| "loss": 0.2596, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_loss": 0.32815971970558167, | |
| "eval_runtime": 64.2353, | |
| "eval_samples_per_second": 3.114, | |
| "eval_steps_per_second": 1.557, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 1.620568037033081, | |
| "learning_rate": 7e-05, | |
| "loss": 0.2272, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 2.3302671909332275, | |
| "learning_rate": 6e-05, | |
| "loss": 0.1611, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 0.3327696919441223, | |
| "eval_runtime": 64.397, | |
| "eval_samples_per_second": 3.106, | |
| "eval_steps_per_second": 1.553, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.511664867401123, | |
| "learning_rate": 5e-05, | |
| "loss": 0.1351, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 3.210343837738037, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1436, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_loss": 0.3513491749763489, | |
| "eval_runtime": 64.5251, | |
| "eval_samples_per_second": 3.1, | |
| "eval_steps_per_second": 1.55, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.3599999999999999, | |
| "grad_norm": 2.1352834701538086, | |
| "learning_rate": 3e-05, | |
| "loss": 0.1259, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 2.6668131351470947, | |
| "learning_rate": 2e-05, | |
| "loss": 0.1498, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "eval_loss": 0.3499194383621216, | |
| "eval_runtime": 63.9987, | |
| "eval_samples_per_second": 3.125, | |
| "eval_steps_per_second": 1.563, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.769700288772583, | |
| "learning_rate": 1e-05, | |
| "loss": 0.1438, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 3.0553314685821533, | |
| "learning_rate": 0.0, | |
| "loss": 0.1443, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 0.3474043011665344, | |
| "eval_runtime": 64.2224, | |
| "eval_samples_per_second": 3.114, | |
| "eval_steps_per_second": 1.557, | |
| "step": 100 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7451874544152192.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |