| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.5458422174840085, |
| "eval_steps": 500, |
| "global_step": 300, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 0.874834306538105, |
| "epoch": 0.08528784648187633, |
| "grad_norm": 9.6875, |
| "learning_rate": 1.25e-06, |
| "loss": 0.7767, |
| "mean_token_accuracy": 0.7803937196731567, |
| "num_tokens": 4793533.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 0.8917073860764504, |
| "epoch": 0.17057569296375266, |
| "grad_norm": 3.359375, |
| "learning_rate": 2.6388888888888893e-06, |
| "loss": 0.7228, |
| "mean_token_accuracy": 0.7852477177977562, |
| "num_tokens": 9585570.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 0.9469680681824684, |
| "epoch": 0.255863539445629, |
| "grad_norm": 1.578125, |
| "learning_rate": 4.027777777777779e-06, |
| "loss": 0.6566, |
| "mean_token_accuracy": 0.7953586861491203, |
| "num_tokens": 14379575.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 0.9756244793534279, |
| "epoch": 0.3411513859275053, |
| "grad_norm": 1.078125, |
| "learning_rate": 4.998902091271986e-06, |
| "loss": 0.5995, |
| "mean_token_accuracy": 0.8040727138519287, |
| "num_tokens": 19171868.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 0.9526244938373566, |
| "epoch": 0.42643923240938164, |
| "grad_norm": 1.265625, |
| "learning_rate": 4.979410528710376e-06, |
| "loss": 0.5673, |
| "mean_token_accuracy": 0.8092003226280212, |
| "num_tokens": 23964483.0, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.938190846145153, |
| "epoch": 0.511727078891258, |
| "grad_norm": 0.8984375, |
| "learning_rate": 4.935739834765994e-06, |
| "loss": 0.5505, |
| "mean_token_accuracy": 0.812424935400486, |
| "num_tokens": 28758375.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.923903401196003, |
| "epoch": 0.5970149253731343, |
| "grad_norm": 2.328125, |
| "learning_rate": 4.868315884635479e-06, |
| "loss": 0.5386, |
| "mean_token_accuracy": 0.8148256585001945, |
| "num_tokens": 33553131.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.9177967429161071, |
| "epoch": 0.6823027718550106, |
| "grad_norm": 0.6953125, |
| "learning_rate": 4.7777961945024834e-06, |
| "loss": 0.5326, |
| "mean_token_accuracy": 0.8151497200131417, |
| "num_tokens": 38347839.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.905217319726944, |
| "epoch": 0.767590618336887, |
| "grad_norm": 0.6484375, |
| "learning_rate": 4.665063509461098e-06, |
| "loss": 0.5279, |
| "mean_token_accuracy": 0.8164947971701622, |
| "num_tokens": 43142598.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8973172023892403, |
| "epoch": 0.8528784648187633, |
| "grad_norm": 1.1796875, |
| "learning_rate": 4.531217195014204e-06, |
| "loss": 0.5221, |
| "mean_token_accuracy": 0.8170435220003128, |
| "num_tokens": 47936856.0, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.8845033064484596, |
| "epoch": 0.9381663113006397, |
| "grad_norm": 1.7578125, |
| "learning_rate": 4.377562516096608e-06, |
| "loss": 0.5185, |
| "mean_token_accuracy": 0.817681685090065, |
| "num_tokens": 52729804.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.879098341271684, |
| "epoch": 1.0170575692963753, |
| "grad_norm": 0.99609375, |
| "learning_rate": 4.205597908173555e-06, |
| "loss": 0.5134, |
| "mean_token_accuracy": 0.8188003140526849, |
| "num_tokens": 57163521.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.8732256084680557, |
| "epoch": 1.1023454157782515, |
| "grad_norm": 0.93359375, |
| "learning_rate": 4.0170003645464835e-06, |
| "loss": 0.512, |
| "mean_token_accuracy": 0.8187085539102554, |
| "num_tokens": 61956161.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.8677664533257484, |
| "epoch": 1.187633262260128, |
| "grad_norm": 2.578125, |
| "learning_rate": 3.8136090823685156e-06, |
| "loss": 0.5085, |
| "mean_token_accuracy": 0.8195212766528129, |
| "num_tokens": 66749986.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.8661862865090371, |
| "epoch": 1.2729211087420043, |
| "grad_norm": 1.140625, |
| "learning_rate": 3.5974075268532354e-06, |
| "loss": 0.5094, |
| "mean_token_accuracy": 0.8192538529634475, |
| "num_tokens": 71544876.0, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.8664106726646423, |
| "epoch": 1.3582089552238805, |
| "grad_norm": 0.65625, |
| "learning_rate": 3.3705040885859975e-06, |
| "loss": 0.5066, |
| "mean_token_accuracy": 0.819321508705616, |
| "num_tokens": 76337411.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.8632829263806343, |
| "epoch": 1.443496801705757, |
| "grad_norm": 0.65625, |
| "learning_rate": 3.1351115225670483e-06, |
| "loss": 0.5053, |
| "mean_token_accuracy": 0.8198848024010659, |
| "num_tokens": 81131176.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.863217756152153, |
| "epoch": 1.5287846481876333, |
| "grad_norm": 0.828125, |
| "learning_rate": 2.8935253694962414e-06, |
| "loss": 0.5, |
| "mean_token_accuracy": 0.8213705286383629, |
| "num_tokens": 85924602.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.8621790811419487, |
| "epoch": 1.6140724946695095, |
| "grad_norm": 0.7890625, |
| "learning_rate": 2.648101569734286e-06, |
| "loss": 0.5007, |
| "mean_token_accuracy": 0.8213415816426277, |
| "num_tokens": 90719186.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.8626999795436859, |
| "epoch": 1.6993603411513858, |
| "grad_norm": 0.8828125, |
| "learning_rate": 2.4012334882484554e-06, |
| "loss": 0.5007, |
| "mean_token_accuracy": 0.8213708132505417, |
| "num_tokens": 95511554.0, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.8624779433012009, |
| "epoch": 1.7846481876332623, |
| "grad_norm": 0.58203125, |
| "learning_rate": 2.1553285745947396e-06, |
| "loss": 0.5025, |
| "mean_token_accuracy": 0.8203421711921692, |
| "num_tokens": 100306193.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.8626597106456757, |
| "epoch": 1.8699360341151388, |
| "grad_norm": 0.54296875, |
| "learning_rate": 1.912784885547541e-06, |
| "loss": 0.5014, |
| "mean_token_accuracy": 0.8207233205437661, |
| "num_tokens": 105100195.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.8611496776342392, |
| "epoch": 1.955223880597015, |
| "grad_norm": 0.6640625, |
| "learning_rate": 1.6759676993274805e-06, |
| "loss": 0.5004, |
| "mean_token_accuracy": 0.8207888290286064, |
| "num_tokens": 109893082.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.8603227138519287, |
| "epoch": 2.0341151385927505, |
| "grad_norm": 0.66796875, |
| "learning_rate": 1.447186449484593e-06, |
| "loss": 0.4991, |
| "mean_token_accuracy": 0.82156281052409, |
| "num_tokens": 114327989.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.8592477083206177, |
| "epoch": 2.1194029850746268, |
| "grad_norm": 0.58203125, |
| "learning_rate": 1.2286722033769494e-06, |
| "loss": 0.4982, |
| "mean_token_accuracy": 0.8215733736753463, |
| "num_tokens": 119121337.0, |
| "step": 250 |
| }, |
| { |
| "entropy": 0.8602951526641845, |
| "epoch": 2.204690831556503, |
| "grad_norm": 0.58203125, |
| "learning_rate": 1.0225559048738548e-06, |
| "loss": 0.4991, |
| "mean_token_accuracy": 0.8218414545059204, |
| "num_tokens": 123915454.0, |
| "step": 260 |
| }, |
| { |
| "entropy": 0.8605464026331902, |
| "epoch": 2.2899786780383797, |
| "grad_norm": 0.54296875, |
| "learning_rate": 8.30847593460069e-07, |
| "loss": 0.4996, |
| "mean_token_accuracy": 0.8212988466024399, |
| "num_tokens": 128710053.0, |
| "step": 270 |
| }, |
| { |
| "entropy": 0.8587785944342613, |
| "epoch": 2.375266524520256, |
| "grad_norm": 0.58203125, |
| "learning_rate": 6.554168023956817e-07, |
| "loss": 0.4978, |
| "mean_token_accuracy": 0.8217180848121644, |
| "num_tokens": 133503872.0, |
| "step": 280 |
| }, |
| { |
| "entropy": 0.8599365219473839, |
| "epoch": 2.4605543710021323, |
| "grad_norm": 0.53515625, |
| "learning_rate": 4.97974327088156e-07, |
| "loss": 0.5003, |
| "mean_token_accuracy": 0.821199668943882, |
| "num_tokens": 138297141.0, |
| "step": 290 |
| }, |
| { |
| "entropy": 0.8603109464049339, |
| "epoch": 2.5458422174840085, |
| "grad_norm": 0.5703125, |
| "learning_rate": 3.6005554147077406e-07, |
| "loss": 0.4988, |
| "mean_token_accuracy": 0.821739350259304, |
| "num_tokens": 143089452.0, |
| "step": 300 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 354, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 3.5170704128506593e+18, |
| "train_batch_size": 8, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
