| { |
| "best_metric": 0.8028125762939453, |
| "best_model_checkpoint": "./llama3/26-07-24-Weni-WeniGPT-Agents-Llama3-5.1.24-SFT_Experiment with DPO and Llama3 70b-2_max_steps-32_batch_8_2024-07-26/checkpoint-32", |
| "epoch": 28.444444444444443, |
| "eval_steps": 2, |
| "global_step": 32, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.7777777777777777, |
| "eval_bleu": 0.02170523150063329, |
| "eval_loss": 1.5546669960021973, |
| "eval_rouge1": 0.5658290942329046, |
| "eval_rouge2": 0.371216207944968, |
| "eval_rougeL": 0.5544380522179795, |
| "eval_rougeLsum": 0.557866697701094, |
| "eval_runtime": 8.903, |
| "eval_samples_per_second": 0.899, |
| "eval_steps_per_second": 0.112, |
| "step": 2 |
| }, |
| { |
| "epoch": 3.5555555555555554, |
| "eval_bleu": 0.016226404911743514, |
| "eval_loss": 1.2373424768447876, |
| "eval_rouge1": 0.1880159013464857, |
| "eval_rouge2": 0.110501293225719, |
| "eval_rougeL": 0.16774584781013288, |
| "eval_rougeLsum": 0.18374679825109685, |
| "eval_runtime": 7.2666, |
| "eval_samples_per_second": 1.101, |
| "eval_steps_per_second": 0.138, |
| "step": 4 |
| }, |
| { |
| "epoch": 5.333333333333333, |
| "eval_bleu": 0.023706516954741297, |
| "eval_loss": 1.123119592666626, |
| "eval_rouge1": 0.6910290907363565, |
| "eval_rouge2": 0.4711964975736579, |
| "eval_rougeL": 0.673625245352441, |
| "eval_rougeLsum": 0.6773221910352027, |
| "eval_runtime": 5.1525, |
| "eval_samples_per_second": 1.553, |
| "eval_steps_per_second": 0.194, |
| "step": 6 |
| }, |
| { |
| "epoch": 7.111111111111111, |
| "eval_bleu": 0.02402168983627918, |
| "eval_loss": 1.0689845085144043, |
| "eval_rouge1": 0.6850615792026293, |
| "eval_rouge2": 0.4793755526234388, |
| "eval_rougeL": 0.6702061146380454, |
| "eval_rougeLsum": 0.6743158691399876, |
| "eval_runtime": 3.5693, |
| "eval_samples_per_second": 2.241, |
| "eval_steps_per_second": 0.28, |
| "step": 8 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "grad_norm": 0.07844142016839124, |
| "learning_rate": 0.00015555702330196023, |
| "loss": 1.3804, |
| "step": 10 |
| }, |
| { |
| "epoch": 8.88888888888889, |
| "eval_bleu": 0.023887172871334808, |
| "eval_loss": 1.0223946571350098, |
| "eval_rouge1": 0.6905898370545817, |
| "eval_rouge2": 0.48557691609082443, |
| "eval_rougeL": 0.6761180651666854, |
| "eval_rougeLsum": 0.6790446004304602, |
| "eval_runtime": 3.5677, |
| "eval_samples_per_second": 2.242, |
| "eval_steps_per_second": 0.28, |
| "step": 10 |
| }, |
| { |
| "epoch": 10.666666666666666, |
| "eval_bleu": 0.024572108418943618, |
| "eval_loss": 0.9860392808914185, |
| "eval_rouge1": 0.7026862650772006, |
| "eval_rouge2": 0.5047043820859161, |
| "eval_rougeL": 0.6864786779541874, |
| "eval_rougeLsum": 0.6905060789723064, |
| "eval_runtime": 3.3868, |
| "eval_samples_per_second": 2.362, |
| "eval_steps_per_second": 0.295, |
| "step": 12 |
| }, |
| { |
| "epoch": 12.444444444444445, |
| "eval_bleu": 0.025429922278916713, |
| "eval_loss": 0.9487287998199463, |
| "eval_rouge1": 0.702795109607377, |
| "eval_rouge2": 0.5129806168696595, |
| "eval_rougeL": 0.6895042238397062, |
| "eval_rougeLsum": 0.6947769064530469, |
| "eval_runtime": 5.1904, |
| "eval_samples_per_second": 1.541, |
| "eval_steps_per_second": 0.193, |
| "step": 14 |
| }, |
| { |
| "epoch": 14.222222222222221, |
| "eval_bleu": 0.025784195587265114, |
| "eval_loss": 0.9159524440765381, |
| "eval_rouge1": 0.7050102641077431, |
| "eval_rouge2": 0.5179038464865816, |
| "eval_rougeL": 0.6926453468372398, |
| "eval_rougeLsum": 0.6956999803401138, |
| "eval_runtime": 3.0778, |
| "eval_samples_per_second": 2.599, |
| "eval_steps_per_second": 0.325, |
| "step": 16 |
| }, |
| { |
| "epoch": 16.0, |
| "eval_bleu": 0.026025861429164285, |
| "eval_loss": 0.8864424228668213, |
| "eval_rouge1": 0.705898212199683, |
| "eval_rouge2": 0.5170955130574988, |
| "eval_rougeL": 0.6904453277300169, |
| "eval_rougeLsum": 0.6940204043388014, |
| "eval_runtime": 2.9696, |
| "eval_samples_per_second": 2.694, |
| "eval_steps_per_second": 0.337, |
| "step": 18 |
| }, |
| { |
| "epoch": 17.77777777777778, |
| "grad_norm": 0.07718723263288825, |
| "learning_rate": 6.173165676349103e-05, |
| "loss": 0.9654, |
| "step": 20 |
| }, |
| { |
| "epoch": 17.77777777777778, |
| "eval_bleu": 0.02596132515691949, |
| "eval_loss": 0.8659763336181641, |
| "eval_rouge1": 0.7122588553468148, |
| "eval_rouge2": 0.5211107654415348, |
| "eval_rougeL": 0.6941911179368457, |
| "eval_rougeLsum": 0.6984106609696767, |
| "eval_runtime": 4.8517, |
| "eval_samples_per_second": 1.649, |
| "eval_steps_per_second": 0.206, |
| "step": 20 |
| }, |
| { |
| "epoch": 19.555555555555557, |
| "eval_bleu": 0.026343897948467077, |
| "eval_loss": 0.8515705466270447, |
| "eval_rouge1": 0.7170480778501949, |
| "eval_rouge2": 0.5292045209358929, |
| "eval_rougeL": 0.6998572077319152, |
| "eval_rougeLsum": 0.7024199215687508, |
| "eval_runtime": 3.1988, |
| "eval_samples_per_second": 2.501, |
| "eval_steps_per_second": 0.313, |
| "step": 22 |
| }, |
| { |
| "epoch": 21.333333333333332, |
| "eval_bleu": 0.025984446331357137, |
| "eval_loss": 0.831386923789978, |
| "eval_rouge1": 0.7153236968867391, |
| "eval_rouge2": 0.5193671791233874, |
| "eval_rougeL": 0.6965103363468828, |
| "eval_rougeLsum": 0.6985595214525613, |
| "eval_runtime": 3.4937, |
| "eval_samples_per_second": 2.29, |
| "eval_steps_per_second": 0.286, |
| "step": 24 |
| }, |
| { |
| "epoch": 23.11111111111111, |
| "eval_bleu": 0.02650924667302787, |
| "eval_loss": 0.8183804750442505, |
| "eval_rouge1": 0.7231112511482293, |
| "eval_rouge2": 0.5345562806445456, |
| "eval_rougeL": 0.7030100294534072, |
| "eval_rougeLsum": 0.7054157022984199, |
| "eval_runtime": 6.0566, |
| "eval_samples_per_second": 1.321, |
| "eval_steps_per_second": 0.165, |
| "step": 26 |
| }, |
| { |
| "epoch": 24.88888888888889, |
| "eval_bleu": 0.02615320831075556, |
| "eval_loss": 0.8091402649879456, |
| "eval_rouge1": 0.7207310641325595, |
| "eval_rouge2": 0.5306337690270762, |
| "eval_rougeL": 0.7021914980969189, |
| "eval_rougeLsum": 0.70505062098815, |
| "eval_runtime": 3.5767, |
| "eval_samples_per_second": 2.237, |
| "eval_steps_per_second": 0.28, |
| "step": 28 |
| }, |
| { |
| "epoch": 26.666666666666668, |
| "grad_norm": 0.08878053905359502, |
| "learning_rate": 1.921471959676957e-06, |
| "loss": 0.7862, |
| "step": 30 |
| }, |
| { |
| "epoch": 26.666666666666668, |
| "eval_bleu": 0.026398236337588678, |
| "eval_loss": 0.8041359186172485, |
| "eval_rouge1": 0.72570713977056, |
| "eval_rouge2": 0.5397446508715937, |
| "eval_rougeL": 0.7070571025409038, |
| "eval_rougeLsum": 0.7093496693847328, |
| "eval_runtime": 3.4925, |
| "eval_samples_per_second": 2.291, |
| "eval_steps_per_second": 0.286, |
| "step": 30 |
| }, |
| { |
| "epoch": 28.444444444444443, |
| "eval_bleu": 0.026431275243775837, |
| "eval_loss": 0.8028125762939453, |
| "eval_rouge1": 0.7266203046783699, |
| "eval_rouge2": 0.5395778050172039, |
| "eval_rougeL": 0.7073728737550329, |
| "eval_rougeLsum": 0.7103352101315058, |
| "eval_runtime": 4.8603, |
| "eval_samples_per_second": 1.646, |
| "eval_steps_per_second": 0.206, |
| "step": 32 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 32, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 32, |
| "save_steps": 2, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 23047781220352.0, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
