| { | |
| "best_metric": 6.85207986831665, | |
| "best_model_checkpoint": "./results/models/mistral-dna/checkpoint-18333", | |
| "epoch": 7.0, | |
| "eval_steps": 500, | |
| "global_step": 18333, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19091256204658266, | |
| "grad_norm": 0.0634765625, | |
| "learning_rate": 0.003984726995036274, | |
| "loss": 6.9779, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.3818251240931653, | |
| "grad_norm": 0.09716796875, | |
| "learning_rate": 0.003969453990072547, | |
| "loss": 6.8377, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.572737686139748, | |
| "grad_norm": 0.07177734375, | |
| "learning_rate": 0.003954180985108821, | |
| "loss": 6.7842, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7636502481863306, | |
| "grad_norm": 0.1025390625, | |
| "learning_rate": 0.003938907980145094, | |
| "loss": 6.8022, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.9545628102329133, | |
| "grad_norm": 1.4921875, | |
| "learning_rate": 0.003923634975181367, | |
| "loss": 6.8364, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 6.887508392333984, | |
| "eval_runtime": 0.7694, | |
| "eval_samples_per_second": 109.176, | |
| "eval_steps_per_second": 3.899, | |
| "step": 2619 | |
| }, | |
| { | |
| "epoch": 1.145475372279496, | |
| "grad_norm": 1.515625, | |
| "learning_rate": 0.0039083619702176406, | |
| "loss": 6.8965, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3363879343260787, | |
| "grad_norm": 1.859375, | |
| "learning_rate": 0.003893088965253914, | |
| "loss": 6.9056, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5273004963726613, | |
| "grad_norm": 2.28125, | |
| "learning_rate": 0.003877815960290187, | |
| "loss": 6.8995, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.718213058419244, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 0.003862542955326461, | |
| "loss": 6.8937, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.9091256204658267, | |
| "grad_norm": 3.859375, | |
| "learning_rate": 0.0038472699503627338, | |
| "loss": 6.8934, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 6.88522481918335, | |
| "eval_runtime": 0.7793, | |
| "eval_samples_per_second": 107.783, | |
| "eval_steps_per_second": 3.849, | |
| "step": 5238 | |
| }, | |
| { | |
| "epoch": 2.1000381825124093, | |
| "grad_norm": 2.953125, | |
| "learning_rate": 0.003831996945399007, | |
| "loss": 6.8913, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.290950744558992, | |
| "grad_norm": 8.75, | |
| "learning_rate": 0.003816723940435281, | |
| "loss": 6.8984, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.4818633066055744, | |
| "grad_norm": 5.96875, | |
| "learning_rate": 0.003801450935471554, | |
| "loss": 6.8961, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.6727758686521574, | |
| "grad_norm": 4.375, | |
| "learning_rate": 0.003786177930507828, | |
| "loss": 6.8904, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.86368843069874, | |
| "grad_norm": 4.15625, | |
| "learning_rate": 0.003770904925544101, | |
| "loss": 6.8954, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 6.886116981506348, | |
| "eval_runtime": 0.7695, | |
| "eval_samples_per_second": 109.165, | |
| "eval_steps_per_second": 3.899, | |
| "step": 7857 | |
| }, | |
| { | |
| "epoch": 3.0546009927453226, | |
| "grad_norm": 3.296875, | |
| "learning_rate": 0.0037556319205803742, | |
| "loss": 6.891, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.245513554791905, | |
| "grad_norm": 2.9375, | |
| "learning_rate": 0.0037403589156166477, | |
| "loss": 6.8839, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.436426116838488, | |
| "grad_norm": 3.0625, | |
| "learning_rate": 0.003725085910652921, | |
| "loss": 6.8833, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.6273386788850708, | |
| "grad_norm": 8.5, | |
| "learning_rate": 0.0037098129056891945, | |
| "loss": 6.8767, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 3.8182512409316534, | |
| "grad_norm": 3.484375, | |
| "learning_rate": 0.003694539900725468, | |
| "loss": 6.8793, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 6.874268531799316, | |
| "eval_runtime": 0.7764, | |
| "eval_samples_per_second": 108.196, | |
| "eval_steps_per_second": 3.864, | |
| "step": 10476 | |
| }, | |
| { | |
| "epoch": 4.009163802978236, | |
| "grad_norm": 8.875, | |
| "learning_rate": 0.003679266895761741, | |
| "loss": 6.8786, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.2000763650248185, | |
| "grad_norm": 3.71875, | |
| "learning_rate": 0.0036639938907980147, | |
| "loss": 6.874, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.390988927071401, | |
| "grad_norm": 5.5, | |
| "learning_rate": 0.003648720885834288, | |
| "loss": 6.8756, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.581901489117984, | |
| "grad_norm": 19.5, | |
| "learning_rate": 0.0036334478808705615, | |
| "loss": 6.8751, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 4.772814051164566, | |
| "grad_norm": 5.84375, | |
| "learning_rate": 0.003618174875906835, | |
| "loss": 6.877, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 4.963726613211149, | |
| "grad_norm": 3.46875, | |
| "learning_rate": 0.003602901870943108, | |
| "loss": 6.8734, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 6.8629889488220215, | |
| "eval_runtime": 0.767, | |
| "eval_samples_per_second": 109.512, | |
| "eval_steps_per_second": 3.911, | |
| "step": 13095 | |
| }, | |
| { | |
| "epoch": 5.154639175257732, | |
| "grad_norm": 5.40625, | |
| "learning_rate": 0.0035876288659793818, | |
| "loss": 6.8751, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.345551737304315, | |
| "grad_norm": 4.78125, | |
| "learning_rate": 0.0035723558610156548, | |
| "loss": 6.8771, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 5.5364642993508975, | |
| "grad_norm": 4.8125, | |
| "learning_rate": 0.003557082856051928, | |
| "loss": 6.8746, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 5.72737686139748, | |
| "grad_norm": 5.875, | |
| "learning_rate": 0.0035418098510882016, | |
| "loss": 6.8753, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 5.918289423444063, | |
| "grad_norm": 5.3125, | |
| "learning_rate": 0.003526536846124475, | |
| "loss": 6.8791, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 6.864416122436523, | |
| "eval_runtime": 0.7754, | |
| "eval_samples_per_second": 108.328, | |
| "eval_steps_per_second": 3.869, | |
| "step": 15714 | |
| }, | |
| { | |
| "epoch": 6.109201985490645, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 0.0035112638411607484, | |
| "loss": 6.8699, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.300114547537228, | |
| "grad_norm": 4.375, | |
| "learning_rate": 0.003495990836197022, | |
| "loss": 6.8667, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 6.49102710958381, | |
| "grad_norm": 11.5, | |
| "learning_rate": 0.0034807178312332952, | |
| "loss": 6.8655, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 6.681939671630393, | |
| "grad_norm": 6.125, | |
| "learning_rate": 0.0034654448262695686, | |
| "loss": 6.8647, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 6.872852233676976, | |
| "grad_norm": 5.9375, | |
| "learning_rate": 0.003450171821305842, | |
| "loss": 6.8588, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 6.85207986831665, | |
| "eval_runtime": 0.768, | |
| "eval_samples_per_second": 109.374, | |
| "eval_steps_per_second": 3.906, | |
| "step": 18333 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 130950, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 3, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.6896228473744097e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |