| { | |
| "best_metric": 0.16516298, | |
| "best_model_checkpoint": "/home/patrickbarker/output/qwen2-vl-7b-instruct/v3-20241215-182208/checkpoint-93", | |
| "epoch": 3.0, | |
| "eval_steps": 200, | |
| "global_step": 93, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "acc": 0.84791154, | |
| "epoch": 0.03225806451612903, | |
| "grad_norm": 27.065223316478477, | |
| "learning_rate": 0.0, | |
| "loss": 1.10534477, | |
| "memory(GiB)": 55.66, | |
| "step": 1, | |
| "train_speed(iter/s)": 0.022247 | |
| }, | |
| { | |
| "acc": 0.83733273, | |
| "epoch": 0.16129032258064516, | |
| "grad_norm": 17.03496531553331, | |
| "learning_rate": 1e-05, | |
| "loss": 1.06814337, | |
| "memory(GiB)": 76.03, | |
| "step": 5, | |
| "train_speed(iter/s)": 0.02886 | |
| }, | |
| { | |
| "acc": 0.91679258, | |
| "epoch": 0.3225806451612903, | |
| "grad_norm": 7.623726487239116, | |
| "learning_rate": 9.920563966078412e-06, | |
| "loss": 0.49230399, | |
| "memory(GiB)": 73.79, | |
| "step": 10, | |
| "train_speed(iter/s)": 0.030297 | |
| }, | |
| { | |
| "acc": 0.92166452, | |
| "epoch": 0.4838709677419355, | |
| "grad_norm": 3.6617099092671213, | |
| "learning_rate": 9.68478015013631e-06, | |
| "loss": 0.49300199, | |
| "memory(GiB)": 73.79, | |
| "step": 15, | |
| "train_speed(iter/s)": 0.030869 | |
| }, | |
| { | |
| "acc": 0.93338518, | |
| "epoch": 0.6451612903225806, | |
| "grad_norm": 3.119220979880785, | |
| "learning_rate": 9.3001411939184e-06, | |
| "loss": 0.41259723, | |
| "memory(GiB)": 73.79, | |
| "step": 20, | |
| "train_speed(iter/s)": 0.031158 | |
| }, | |
| { | |
| "acc": 0.92974329, | |
| "epoch": 0.8064516129032258, | |
| "grad_norm": 4.078533788825603, | |
| "learning_rate": 8.778869996984113e-06, | |
| "loss": 0.38431735, | |
| "memory(GiB)": 73.79, | |
| "step": 25, | |
| "train_speed(iter/s)": 0.031339 | |
| }, | |
| { | |
| "acc": 0.91595602, | |
| "epoch": 0.967741935483871, | |
| "grad_norm": 10.998030685015813, | |
| "learning_rate": 8.13753130240057e-06, | |
| "loss": 0.40506763, | |
| "memory(GiB)": 73.79, | |
| "step": 30, | |
| "train_speed(iter/s)": 0.031457 | |
| }, | |
| { | |
| "acc": 0.9471405, | |
| "epoch": 1.129032258064516, | |
| "grad_norm": 2.6744991236459517, | |
| "learning_rate": 7.396505309106925e-06, | |
| "loss": 0.27681351, | |
| "memory(GiB)": 73.79, | |
| "step": 35, | |
| "train_speed(iter/s)": 0.031512 | |
| }, | |
| { | |
| "acc": 0.932164, | |
| "epoch": 1.2903225806451613, | |
| "grad_norm": 2.113425035726855, | |
| "learning_rate": 6.579340038281318e-06, | |
| "loss": 0.35436311, | |
| "memory(GiB)": 73.79, | |
| "step": 40, | |
| "train_speed(iter/s)": 0.031584 | |
| }, | |
| { | |
| "acc": 0.92934284, | |
| "epoch": 1.4516129032258065, | |
| "grad_norm": 2.6570760550105366, | |
| "learning_rate": 5.712003033947289e-06, | |
| "loss": 0.3613925, | |
| "memory(GiB)": 73.79, | |
| "step": 45, | |
| "train_speed(iter/s)": 0.031638 | |
| }, | |
| { | |
| "acc": 0.93117428, | |
| "epoch": 1.6129032258064515, | |
| "grad_norm": 0.9540185420283845, | |
| "learning_rate": 4.822056176972018e-06, | |
| "loss": 0.37923932, | |
| "memory(GiB)": 73.79, | |
| "step": 50, | |
| "train_speed(iter/s)": 0.031682 | |
| }, | |
| { | |
| "acc": 0.93338251, | |
| "epoch": 1.7741935483870968, | |
| "grad_norm": 1.4962776432309268, | |
| "learning_rate": 3.937779834879894e-06, | |
| "loss": 0.37198489, | |
| "memory(GiB)": 73.79, | |
| "step": 55, | |
| "train_speed(iter/s)": 0.031717 | |
| }, | |
| { | |
| "acc": 0.93398943, | |
| "epoch": 1.935483870967742, | |
| "grad_norm": 1.8156212498163269, | |
| "learning_rate": 3.0872741798907336e-06, | |
| "loss": 0.36215272, | |
| "memory(GiB)": 73.79, | |
| "step": 60, | |
| "train_speed(iter/s)": 0.031744 | |
| }, | |
| { | |
| "acc": 0.95252762, | |
| "epoch": 2.096774193548387, | |
| "grad_norm": 1.6592309190503378, | |
| "learning_rate": 2.2975662331307417e-06, | |
| "loss": 0.2656436, | |
| "memory(GiB)": 73.79, | |
| "step": 65, | |
| "train_speed(iter/s)": 0.031749 | |
| }, | |
| { | |
| "acc": 0.93537579, | |
| "epoch": 2.258064516129032, | |
| "grad_norm": 1.698659282567525, | |
| "learning_rate": 1.5937510110003426e-06, | |
| "loss": 0.31273713, | |
| "memory(GiB)": 73.79, | |
| "step": 70, | |
| "train_speed(iter/s)": 0.031772 | |
| }, | |
| { | |
| "acc": 0.94613972, | |
| "epoch": 2.4193548387096775, | |
| "grad_norm": 1.5348577615781052, | |
| "learning_rate": 9.981940659986607e-07, | |
| "loss": 0.28420959, | |
| "memory(GiB)": 73.79, | |
| "step": 75, | |
| "train_speed(iter/s)": 0.031793 | |
| }, | |
| { | |
| "acc": 0.93988018, | |
| "epoch": 2.5806451612903225, | |
| "grad_norm": 1.7810683323628338, | |
| "learning_rate": 5.298207633363734e-07, | |
| "loss": 0.30640914, | |
| "memory(GiB)": 73.79, | |
| "step": 80, | |
| "train_speed(iter/s)": 0.031807 | |
| }, | |
| { | |
| "acc": 0.93680553, | |
| "epoch": 2.741935483870968, | |
| "grad_norm": 2.5905151386873873, | |
| "learning_rate": 2.035148784143204e-07, | |
| "loss": 0.33139644, | |
| "memory(GiB)": 73.79, | |
| "step": 85, | |
| "train_speed(iter/s)": 0.031819 | |
| }, | |
| { | |
| "acc": 0.94427986, | |
| "epoch": 2.903225806451613, | |
| "grad_norm": 1.829616630873348, | |
| "learning_rate": 2.9645626291399768e-08, | |
| "loss": 0.32220521, | |
| "memory(GiB)": 73.79, | |
| "step": 90, | |
| "train_speed(iter/s)": 0.03183 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_acc": 0.9263189812007278, | |
| "eval_loss": 0.16516298055648804, | |
| "eval_runtime": 76.0876, | |
| "eval_samples_per_second": 1.603, | |
| "eval_steps_per_second": 0.21, | |
| "step": 93 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 93, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 216369601708032.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |