LMMs-Lab-Turtle/Qwen-2.5VL-3B-Cold-Start
07
1{2 "best_metric": null,3 "best_model_checkpoint": null,4 "epoch": 2.957983193277311,5 "eval_steps": 500,6 "global_step": 177,7 "is_hyper_param_search": false,8 "is_local_process_zero": true,9 "is_world_process_zero": true,10 "log_history": [11 {12 "epoch": 0.16806722689075632,13 "grad_norm": 1.64210754465846,14 "learning_rate": 5.555555555555557e-06,15 "loss": 0.8834,16 "step": 1017 },18 {19 "epoch": 0.33613445378151263,20 "grad_norm": 1.6216633128454179,21 "learning_rate": 9.99609654676786e-06,22 "loss": 0.832,23 "step": 2024 },25 {26 "epoch": 0.5042016806722689,27 "grad_norm": 1.0890026008565015,28 "learning_rate": 9.860114570402055e-06,29 "loss": 0.7929,30 "step": 3031 },32 {33 "epoch": 0.6722689075630253,34 "grad_norm": 0.9852931548778409,35 "learning_rate": 9.535012074008688e-06,36 "loss": 0.7613,37 "step": 4038 },39 {40 "epoch": 0.8403361344537815,41 "grad_norm": 1.0010829198190254,42 "learning_rate": 9.033439696227966e-06,43 "loss": 0.7369,44 "step": 5045 },46 {47 "epoch": 1.0,48 "grad_norm": 0.9912119135873784,49 "learning_rate": 8.374915007591053e-06,50 "loss": 0.6853,51 "step": 6052 },53 {54 "epoch": 1.1680672268907564,55 "grad_norm": 0.9598316653212201,56 "learning_rate": 7.58506302778873e-06,57 "loss": 0.6288,58 "step": 7059 },60 {61 "epoch": 1.3361344537815127,62 "grad_norm": 0.9969526888332303,63 "learning_rate": 6.694619085176159e-06,64 "loss": 0.5971,65 "step": 8066 },67 {68 "epoch": 1.504201680672269,69 "grad_norm": 5.712426354685314,70 "learning_rate": 5.738232820012407e-06,71 "loss": 0.5633,72 "step": 9073 },74 {75 "epoch": 1.6722689075630253,76 "grad_norm": 0.8053877541293872,77 "learning_rate": 4.753119870981486e-06,78 "loss": 0.5586,79 "step": 10080 },81 {82 "epoch": 1.8403361344537816,83 "grad_norm": 0.9085557253419428,84 "learning_rate": 3.777613711607087e-06,85 "loss": 0.5633,86 "step": 11087 },88 {89 "epoch": 2.0,90 "grad_norm": 0.9125623622985397,91 "learning_rate": 2.8496739886173994e-06,92 "loss": 0.5249,93 "step": 12094 },95 {96 "epoch": 2.168067226890756,97 "grad_norm": 0.9796403145730364,98 "learning_rate": 2.005409406946e-06,99 "loss": 0.4774,100 "step": 130101 },102 {103 "epoch": 2.3361344537815127,104 "grad_norm": 0.8379009664030831,105 "learning_rate": 1.277672640004936e-06,106 "loss": 0.4569,107 "step": 140108 },109 {110 "epoch": 2.504201680672269,111 "grad_norm": 0.912579146579012,112 "learning_rate": 6.947819411632223e-07,113 "loss": 0.4598,114 "step": 150115 },116 {117 "epoch": 2.6722689075630255,118 "grad_norm": 0.8570497984949336,119 "learning_rate": 2.7941920206915443e-07,120 "loss": 0.4533,121 "step": 160122 },123 {124 "epoch": 2.8403361344537816,125 "grad_norm": 0.913025927962694,126 "learning_rate": 4.774733741942206e-08,127 "loss": 0.4608,128 "step": 170129 },130 {131 "epoch": 2.957983193277311,132 "step": 177,133 "total_flos": 92447203917824.0,134 "train_loss": 0.6085002004763501,135 "train_runtime": 1135.371,136 "train_samples_per_second": 20.124,137 "train_steps_per_second": 0.156138 }139 ],140 "logging_steps": 10,141 "max_steps": 177,142 "num_input_tokens_seen": 0,143 "num_train_epochs": 3,144 "save_steps": 500,145 "stateful_callbacks": {146 "TrainerControl": {147 "args": {148 "should_epoch_stop": false,149 "should_evaluate": false,150 "should_log": false,151 "should_save": true,152 "should_training_stop": true153 },154 "attributes": {}155 }156 },157 "total_flos": 92447203917824.0,158 "train_batch_size": 8,159 "trial_name": null,160 "trial_params": null161}162 