| { |
| "best_global_step": 275, |
| "best_metric": 0.9421520233154297, |
| "best_model_checkpoint": "./dspy-8bit-lora-deepcoder-safe/checkpoint-275", |
| "epoch": 2.9982300884955753, |
| "eval_steps": 25, |
| "global_step": 425, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.17699115044247787, |
| "grad_norm": 0.6848811507225037, |
| "learning_rate": 4.8e-05, |
| "loss": 1.9452, |
| "mean_token_accuracy": 0.5935992772877217, |
| "num_tokens": 53941.0, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.17699115044247787, |
| "eval_loss": 1.9623712301254272, |
| "eval_mean_token_accuracy": 0.6255391545593738, |
| "eval_num_tokens": 53941.0, |
| "eval_runtime": 96.8734, |
| "eval_samples_per_second": 2.065, |
| "eval_steps_per_second": 2.065, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.35398230088495575, |
| "grad_norm": 5.549977779388428, |
| "learning_rate": 9.8e-05, |
| "loss": 1.4238, |
| "mean_token_accuracy": 0.6662441535294056, |
| "num_tokens": 110474.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.35398230088495575, |
| "eval_loss": 1.394215703010559, |
| "eval_mean_token_accuracy": 0.6895178978145122, |
| "eval_num_tokens": 110474.0, |
| "eval_runtime": 96.6623, |
| "eval_samples_per_second": 2.069, |
| "eval_steps_per_second": 2.069, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.5309734513274337, |
| "grad_norm": 0.6720989942550659, |
| "learning_rate": 0.000148, |
| "loss": 1.198, |
| "mean_token_accuracy": 0.7074533948302268, |
| "num_tokens": 162407.0, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.5309734513274337, |
| "eval_loss": 1.203566074371338, |
| "eval_mean_token_accuracy": 0.7197688557207584, |
| "eval_num_tokens": 162407.0, |
| "eval_runtime": 97.3608, |
| "eval_samples_per_second": 2.054, |
| "eval_steps_per_second": 2.054, |
| "step": 75 |
| }, |
| { |
| "epoch": 0.7079646017699115, |
| "grad_norm": 0.5576093792915344, |
| "learning_rate": 0.00019800000000000002, |
| "loss": 1.0969, |
| "mean_token_accuracy": 0.7274327605962754, |
| "num_tokens": 225140.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.7079646017699115, |
| "eval_loss": 1.1339093446731567, |
| "eval_mean_token_accuracy": 0.7314411370456219, |
| "eval_num_tokens": 225140.0, |
| "eval_runtime": 98.1154, |
| "eval_samples_per_second": 2.038, |
| "eval_steps_per_second": 2.038, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.8849557522123894, |
| "grad_norm": 0.5583345293998718, |
| "learning_rate": 0.0001852760736196319, |
| "loss": 1.0369, |
| "mean_token_accuracy": 0.7487157288193703, |
| "num_tokens": 285818.0, |
| "step": 125 |
| }, |
| { |
| "epoch": 0.8849557522123894, |
| "eval_loss": 1.0866804122924805, |
| "eval_mean_token_accuracy": 0.739626374989748, |
| "eval_num_tokens": 285818.0, |
| "eval_runtime": 97.5506, |
| "eval_samples_per_second": 2.05, |
| "eval_steps_per_second": 2.05, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.056637168141593, |
| "grad_norm": 1.0887213945388794, |
| "learning_rate": 0.00016993865030674846, |
| "loss": 0.9094, |
| "mean_token_accuracy": 0.767275901492109, |
| "num_tokens": 338145.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.056637168141593, |
| "eval_loss": 1.091628074645996, |
| "eval_mean_token_accuracy": 0.7472012284398079, |
| "eval_num_tokens": 338145.0, |
| "eval_runtime": 97.9463, |
| "eval_samples_per_second": 2.042, |
| "eval_steps_per_second": 2.042, |
| "step": 150 |
| }, |
| { |
| "epoch": 1.2336283185840708, |
| "grad_norm": 1.2973228693008423, |
| "learning_rate": 0.00015460122699386504, |
| "loss": 0.7983, |
| "mean_token_accuracy": 0.786946276128292, |
| "num_tokens": 395324.0, |
| "step": 175 |
| }, |
| { |
| "epoch": 1.2336283185840708, |
| "eval_loss": 1.0309680700302124, |
| "eval_mean_token_accuracy": 0.7567924955487251, |
| "eval_num_tokens": 395324.0, |
| "eval_runtime": 97.3557, |
| "eval_samples_per_second": 2.054, |
| "eval_steps_per_second": 2.054, |
| "step": 175 |
| }, |
| { |
| "epoch": 1.4106194690265488, |
| "grad_norm": 0.8981689810752869, |
| "learning_rate": 0.00013926380368098159, |
| "loss": 0.8004, |
| "mean_token_accuracy": 0.7892016369104385, |
| "num_tokens": 453392.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.4106194690265488, |
| "eval_loss": 1.0047160387039185, |
| "eval_mean_token_accuracy": 0.7607714273035526, |
| "eval_num_tokens": 453392.0, |
| "eval_runtime": 97.3422, |
| "eval_samples_per_second": 2.055, |
| "eval_steps_per_second": 2.055, |
| "step": 200 |
| }, |
| { |
| "epoch": 1.5876106194690265, |
| "grad_norm": 1.3156253099441528, |
| "learning_rate": 0.00012392638036809816, |
| "loss": 0.7496, |
| "mean_token_accuracy": 0.800215690433979, |
| "num_tokens": 513773.0, |
| "step": 225 |
| }, |
| { |
| "epoch": 1.5876106194690265, |
| "eval_loss": 0.9816662073135376, |
| "eval_mean_token_accuracy": 0.7670722763240337, |
| "eval_num_tokens": 513773.0, |
| "eval_runtime": 98.5747, |
| "eval_samples_per_second": 2.029, |
| "eval_steps_per_second": 2.029, |
| "step": 225 |
| }, |
| { |
| "epoch": 1.7646017699115044, |
| "grad_norm": 0.7007508277893066, |
| "learning_rate": 0.00010858895705521473, |
| "loss": 0.7315, |
| "mean_token_accuracy": 0.805658475458622, |
| "num_tokens": 571382.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.7646017699115044, |
| "eval_loss": 0.95888352394104, |
| "eval_mean_token_accuracy": 0.7721423482894898, |
| "eval_num_tokens": 571382.0, |
| "eval_runtime": 97.9779, |
| "eval_samples_per_second": 2.041, |
| "eval_steps_per_second": 2.041, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.9415929203539823, |
| "grad_norm": 0.8127393126487732, |
| "learning_rate": 9.325153374233129e-05, |
| "loss": 0.7173, |
| "mean_token_accuracy": 0.8105942443013191, |
| "num_tokens": 628449.0, |
| "step": 275 |
| }, |
| { |
| "epoch": 1.9415929203539823, |
| "eval_loss": 0.9421520233154297, |
| "eval_mean_token_accuracy": 0.7754264496266842, |
| "eval_num_tokens": 628449.0, |
| "eval_runtime": 97.1509, |
| "eval_samples_per_second": 2.059, |
| "eval_steps_per_second": 2.059, |
| "step": 275 |
| }, |
| { |
| "epoch": 2.113274336283186, |
| "grad_norm": 1.1352859735488892, |
| "learning_rate": 7.791411042944787e-05, |
| "loss": 0.5105, |
| "mean_token_accuracy": 0.8593325974400511, |
| "num_tokens": 680687.0, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.113274336283186, |
| "eval_loss": 0.9843020439147949, |
| "eval_mean_token_accuracy": 0.7742766647040844, |
| "eval_num_tokens": 680687.0, |
| "eval_runtime": 98.2979, |
| "eval_samples_per_second": 2.035, |
| "eval_steps_per_second": 2.035, |
| "step": 300 |
| }, |
| { |
| "epoch": 2.2902654867256635, |
| "grad_norm": 0.8646391034126282, |
| "learning_rate": 6.257668711656443e-05, |
| "loss": 0.4312, |
| "mean_token_accuracy": 0.878243299126625, |
| "num_tokens": 733979.0, |
| "step": 325 |
| }, |
| { |
| "epoch": 2.2902654867256635, |
| "eval_loss": 0.993569552898407, |
| "eval_mean_token_accuracy": 0.775199833214283, |
| "eval_num_tokens": 733979.0, |
| "eval_runtime": 96.9581, |
| "eval_samples_per_second": 2.063, |
| "eval_steps_per_second": 2.063, |
| "step": 325 |
| }, |
| { |
| "epoch": 2.4672566371681417, |
| "grad_norm": 1.0304445028305054, |
| "learning_rate": 4.723926380368098e-05, |
| "loss": 0.3952, |
| "mean_token_accuracy": 0.8835302656888961, |
| "num_tokens": 789649.0, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.4672566371681417, |
| "eval_loss": 0.9702684283256531, |
| "eval_mean_token_accuracy": 0.77936582878232, |
| "eval_num_tokens": 789649.0, |
| "eval_runtime": 98.4219, |
| "eval_samples_per_second": 2.032, |
| "eval_steps_per_second": 2.032, |
| "step": 350 |
| }, |
| { |
| "epoch": 2.6442477876106194, |
| "grad_norm": 0.8441785573959351, |
| "learning_rate": 3.1901840490797544e-05, |
| "loss": 0.3966, |
| "mean_token_accuracy": 0.8844721549749375, |
| "num_tokens": 845295.0, |
| "step": 375 |
| }, |
| { |
| "epoch": 2.6442477876106194, |
| "eval_loss": 0.9620086550712585, |
| "eval_mean_token_accuracy": 0.7812778241932392, |
| "eval_num_tokens": 845295.0, |
| "eval_runtime": 97.2727, |
| "eval_samples_per_second": 2.056, |
| "eval_steps_per_second": 2.056, |
| "step": 375 |
| }, |
| { |
| "epoch": 2.8212389380530976, |
| "grad_norm": 0.8250017762184143, |
| "learning_rate": 1.656441717791411e-05, |
| "loss": 0.3743, |
| "mean_token_accuracy": 0.8831119546294213, |
| "num_tokens": 908151.0, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.8212389380530976, |
| "eval_loss": 0.9527075290679932, |
| "eval_mean_token_accuracy": 0.7827157293260097, |
| "eval_num_tokens": 908151.0, |
| "eval_runtime": 98.1465, |
| "eval_samples_per_second": 2.038, |
| "eval_steps_per_second": 2.038, |
| "step": 400 |
| }, |
| { |
| "epoch": 2.9982300884955753, |
| "grad_norm": 0.8464552164077759, |
| "learning_rate": 1.226993865030675e-06, |
| "loss": 0.4002, |
| "mean_token_accuracy": 0.883520859181881, |
| "num_tokens": 967441.0, |
| "step": 425 |
| }, |
| { |
| "epoch": 2.9982300884955753, |
| "eval_loss": 0.9509771466255188, |
| "eval_mean_token_accuracy": 0.7838174617290496, |
| "eval_num_tokens": 967441.0, |
| "eval_runtime": 98.6966, |
| "eval_samples_per_second": 2.026, |
| "eval_steps_per_second": 2.026, |
| "step": 425 |
| } |
| ], |
| "logging_steps": 25, |
| "max_steps": 426, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 25, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 8.204357166595891e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
