| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 5.0, |
| "eval_steps": 500, |
| "global_step": 14910, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.1676727028839705, |
| "grad_norm": 0.5760945677757263, |
| "learning_rate": 4.83232729711603e-05, |
| "loss": 0.2766, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.335345405767941, |
| "grad_norm": 0.3646267056465149, |
| "learning_rate": 4.6646545942320594e-05, |
| "loss": 0.2008, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.5030181086519114, |
| "grad_norm": 0.3581697344779968, |
| "learning_rate": 4.4969818913480886e-05, |
| "loss": 0.1872, |
| "step": 1500 |
| }, |
| { |
| "epoch": 0.670690811535882, |
| "grad_norm": 0.26507869362831116, |
| "learning_rate": 4.3293091884641186e-05, |
| "loss": 0.164, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.8383635144198525, |
| "grad_norm": 0.2440381944179535, |
| "learning_rate": 4.161636485580148e-05, |
| "loss": 0.1504, |
| "step": 2500 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_gen_len": 100.0, |
| "eval_loss": 0.10272549837827682, |
| "eval_rouge1": 0.2262, |
| "eval_rouge2": 0.2262, |
| "eval_rougeL": 0.2262, |
| "eval_rougeLsum": 0.2262, |
| "eval_runtime": 1543.4357, |
| "eval_samples_per_second": 0.859, |
| "eval_steps_per_second": 0.215, |
| "step": 2982 |
| }, |
| { |
| "epoch": 1.0060362173038229, |
| "grad_norm": 0.48001933097839355, |
| "learning_rate": 3.993963782696177e-05, |
| "loss": 0.1343, |
| "step": 3000 |
| }, |
| { |
| "epoch": 1.1737089201877935, |
| "grad_norm": 0.47931984066963196, |
| "learning_rate": 3.826291079812207e-05, |
| "loss": 0.0978, |
| "step": 3500 |
| }, |
| { |
| "epoch": 1.341381623071764, |
| "grad_norm": 0.42881467938423157, |
| "learning_rate": 3.658618376928236e-05, |
| "loss": 0.0866, |
| "step": 4000 |
| }, |
| { |
| "epoch": 1.5090543259557343, |
| "grad_norm": 0.4142566919326782, |
| "learning_rate": 3.490945674044266e-05, |
| "loss": 0.0882, |
| "step": 4500 |
| }, |
| { |
| "epoch": 1.676727028839705, |
| "grad_norm": 0.13182227313518524, |
| "learning_rate": 3.323272971160295e-05, |
| "loss": 0.0759, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.8443997317236756, |
| "grad_norm": 0.3671654164791107, |
| "learning_rate": 3.1556002682763246e-05, |
| "loss": 0.0664, |
| "step": 5500 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_gen_len": 100.0, |
| "eval_loss": 0.045729078352451324, |
| "eval_rouge1": 1.7181, |
| "eval_rouge2": 1.2785, |
| "eval_rougeL": 1.7523, |
| "eval_rougeLsum": 1.7256, |
| "eval_runtime": 1544.7805, |
| "eval_samples_per_second": 0.858, |
| "eval_steps_per_second": 0.215, |
| "step": 5964 |
| }, |
| { |
| "epoch": 2.0120724346076457, |
| "grad_norm": 0.25756004452705383, |
| "learning_rate": 2.9879275653923545e-05, |
| "loss": 0.0643, |
| "step": 6000 |
| }, |
| { |
| "epoch": 2.1797451374916164, |
| "grad_norm": 0.3812272548675537, |
| "learning_rate": 2.8202548625083835e-05, |
| "loss": 0.0392, |
| "step": 6500 |
| }, |
| { |
| "epoch": 2.347417840375587, |
| "grad_norm": 0.38306811451911926, |
| "learning_rate": 2.6525821596244134e-05, |
| "loss": 0.0373, |
| "step": 7000 |
| }, |
| { |
| "epoch": 2.515090543259557, |
| "grad_norm": 0.4030314087867737, |
| "learning_rate": 2.4849094567404426e-05, |
| "loss": 0.0372, |
| "step": 7500 |
| }, |
| { |
| "epoch": 2.682763246143528, |
| "grad_norm": 0.309796541929245, |
| "learning_rate": 2.3172367538564725e-05, |
| "loss": 0.0324, |
| "step": 8000 |
| }, |
| { |
| "epoch": 2.8504359490274984, |
| "grad_norm": 0.19243386387825012, |
| "learning_rate": 2.1495640509725018e-05, |
| "loss": 0.0281, |
| "step": 8500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_gen_len": 100.0, |
| "eval_loss": 0.0184603463858366, |
| "eval_rouge1": 5.3843, |
| "eval_rouge2": 3.3178, |
| "eval_rougeL": 5.4006, |
| "eval_rougeLsum": 5.4032, |
| "eval_runtime": 1543.3574, |
| "eval_samples_per_second": 0.859, |
| "eval_steps_per_second": 0.215, |
| "step": 8946 |
| }, |
| { |
| "epoch": 3.0181086519114686, |
| "grad_norm": 0.09871068596839905, |
| "learning_rate": 1.9818913480885314e-05, |
| "loss": 0.0248, |
| "step": 9000 |
| }, |
| { |
| "epoch": 3.1857813547954392, |
| "grad_norm": 0.6722275614738464, |
| "learning_rate": 1.814218645204561e-05, |
| "loss": 0.018, |
| "step": 9500 |
| }, |
| { |
| "epoch": 3.35345405767941, |
| "grad_norm": 0.2110537886619568, |
| "learning_rate": 1.6465459423205902e-05, |
| "loss": 0.0166, |
| "step": 10000 |
| }, |
| { |
| "epoch": 3.52112676056338, |
| "grad_norm": 0.2037188857793808, |
| "learning_rate": 1.4788732394366198e-05, |
| "loss": 0.0131, |
| "step": 10500 |
| }, |
| { |
| "epoch": 3.6887994634473507, |
| "grad_norm": 0.46498414874076843, |
| "learning_rate": 1.3112005365526492e-05, |
| "loss": 0.0142, |
| "step": 11000 |
| }, |
| { |
| "epoch": 3.8564721663313213, |
| "grad_norm": 0.1339973509311676, |
| "learning_rate": 1.1435278336686788e-05, |
| "loss": 0.0131, |
| "step": 11500 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_gen_len": 100.0, |
| "eval_loss": 0.008116796612739563, |
| "eval_rouge1": 10.8352, |
| "eval_rouge2": 5.9986, |
| "eval_rougeL": 10.7776, |
| "eval_rougeLsum": 10.8786, |
| "eval_runtime": 1543.4295, |
| "eval_samples_per_second": 0.859, |
| "eval_steps_per_second": 0.215, |
| "step": 11928 |
| }, |
| { |
| "epoch": 4.0241448692152915, |
| "grad_norm": 0.11288804560899734, |
| "learning_rate": 9.758551307847083e-06, |
| "loss": 0.0113, |
| "step": 12000 |
| }, |
| { |
| "epoch": 4.1918175720992625, |
| "grad_norm": 0.18565210700035095, |
| "learning_rate": 8.08182427900738e-06, |
| "loss": 0.0088, |
| "step": 12500 |
| }, |
| { |
| "epoch": 4.359490274983233, |
| "grad_norm": 0.2599884867668152, |
| "learning_rate": 6.4050972501676726e-06, |
| "loss": 0.0083, |
| "step": 13000 |
| }, |
| { |
| "epoch": 4.527162977867203, |
| "grad_norm": 0.11985205858945847, |
| "learning_rate": 4.7283702213279675e-06, |
| "loss": 0.008, |
| "step": 13500 |
| }, |
| { |
| "epoch": 4.694835680751174, |
| "grad_norm": 0.10148580372333527, |
| "learning_rate": 3.051643192488263e-06, |
| "loss": 0.0079, |
| "step": 14000 |
| }, |
| { |
| "epoch": 4.862508383635144, |
| "grad_norm": 0.19774800539016724, |
| "learning_rate": 1.3749161636485581e-06, |
| "loss": 0.007, |
| "step": 14500 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 14910, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 5000, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.0760159508037632e+19, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
