| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 7.917525773195877, | |
| "eval_steps": 30, | |
| "global_step": 48, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16494845360824742, | |
| "grad_norm": 1.603933334350586, | |
| "learning_rate": 2.9375e-05, | |
| "loss": 10.1011, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.32989690721649484, | |
| "grad_norm": 1.7179864645004272, | |
| "learning_rate": 2.875e-05, | |
| "loss": 10.2338, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.4948453608247423, | |
| "grad_norm": 1.7488847970962524, | |
| "learning_rate": 2.8125e-05, | |
| "loss": 10.0559, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.6597938144329897, | |
| "grad_norm": 2.1688830852508545, | |
| "learning_rate": 2.75e-05, | |
| "loss": 10.2016, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8247422680412371, | |
| "grad_norm": 2.111985206604004, | |
| "learning_rate": 2.6875000000000003e-05, | |
| "loss": 10.0667, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.9896907216494846, | |
| "grad_norm": 2.164363145828247, | |
| "learning_rate": 2.625e-05, | |
| "loss": 9.8519, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.1546391752577319, | |
| "grad_norm": 2.308406114578247, | |
| "learning_rate": 2.5625e-05, | |
| "loss": 9.92, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.3195876288659794, | |
| "grad_norm": 2.625577688217163, | |
| "learning_rate": 2.5e-05, | |
| "loss": 9.8473, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.4845360824742269, | |
| "grad_norm": 3.0818679332733154, | |
| "learning_rate": 2.4375e-05, | |
| "loss": 9.8717, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 1.6494845360824741, | |
| "grad_norm": 2.6675806045532227, | |
| "learning_rate": 2.3749999999999998e-05, | |
| "loss": 9.6087, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.8144329896907216, | |
| "grad_norm": 2.8696348667144775, | |
| "learning_rate": 2.3125000000000003e-05, | |
| "loss": 9.6005, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 1.9793814432989691, | |
| "grad_norm": 3.16064453125, | |
| "learning_rate": 2.25e-05, | |
| "loss": 9.5111, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.1443298969072164, | |
| "grad_norm": 3.2917752265930176, | |
| "learning_rate": 2.1875e-05, | |
| "loss": 9.4589, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 2.3092783505154637, | |
| "grad_norm": 3.362398624420166, | |
| "learning_rate": 2.125e-05, | |
| "loss": 9.3264, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.4742268041237114, | |
| "grad_norm": 3.687838315963745, | |
| "learning_rate": 2.0625e-05, | |
| "loss": 9.3032, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.6391752577319587, | |
| "grad_norm": 3.6583521366119385, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 9.1917, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.804123711340206, | |
| "grad_norm": 3.846595048904419, | |
| "learning_rate": 1.9375e-05, | |
| "loss": 9.0754, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.9690721649484537, | |
| "grad_norm": 3.7431483268737793, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 9.0118, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.134020618556701, | |
| "grad_norm": 3.54080867767334, | |
| "learning_rate": 1.8125e-05, | |
| "loss": 8.8953, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 3.2989690721649483, | |
| "grad_norm": 3.989246368408203, | |
| "learning_rate": 1.7500000000000002e-05, | |
| "loss": 8.8344, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.463917525773196, | |
| "grad_norm": 3.540311574935913, | |
| "learning_rate": 1.6875e-05, | |
| "loss": 8.7965, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 3.6288659793814433, | |
| "grad_norm": 3.735725164413452, | |
| "learning_rate": 1.625e-05, | |
| "loss": 8.6344, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 3.7938144329896906, | |
| "grad_norm": 4.285547733306885, | |
| "learning_rate": 1.5625e-05, | |
| "loss": 8.544, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.9587628865979383, | |
| "grad_norm": 3.555814027786255, | |
| "learning_rate": 1.5e-05, | |
| "loss": 8.4752, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.123711340206185, | |
| "grad_norm": 3.49817156791687, | |
| "learning_rate": 1.4375e-05, | |
| "loss": 8.4403, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.288659793814433, | |
| "grad_norm": 3.3520140647888184, | |
| "learning_rate": 1.375e-05, | |
| "loss": 8.4122, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.453608247422681, | |
| "grad_norm": 3.296865224838257, | |
| "learning_rate": 1.3125e-05, | |
| "loss": 8.3044, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 4.618556701030927, | |
| "grad_norm": 3.4501516819000244, | |
| "learning_rate": 1.25e-05, | |
| "loss": 8.1882, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.783505154639175, | |
| "grad_norm": 3.2213199138641357, | |
| "learning_rate": 1.1874999999999999e-05, | |
| "loss": 8.2403, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "grad_norm": 3.34763765335083, | |
| "learning_rate": 1.125e-05, | |
| "loss": 8.1814, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.948453608247423, | |
| "eval_audio_cosine_sim": 0.5797536969184875, | |
| "eval_loss": 3.095458745956421, | |
| "eval_runtime": 2074.7641, | |
| "eval_samples_per_second": 0.006, | |
| "eval_steps_per_second": 0.006, | |
| "eval_text_cosine_sim": 0.38171055912971497, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.11340206185567, | |
| "grad_norm": 3.406193256378174, | |
| "learning_rate": 1.0625e-05, | |
| "loss": 8.109, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 5.278350515463917, | |
| "grad_norm": 3.601793050765991, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 7.9923, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.443298969072165, | |
| "grad_norm": 3.392193555831909, | |
| "learning_rate": 9.375000000000001e-06, | |
| "loss": 8.202, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 5.608247422680412, | |
| "grad_norm": 3.5988409519195557, | |
| "learning_rate": 8.750000000000001e-06, | |
| "loss": 8.0204, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.77319587628866, | |
| "grad_norm": 3.4323904514312744, | |
| "learning_rate": 8.125e-06, | |
| "loss": 8.008, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 5.938144329896907, | |
| "grad_norm": 3.5866267681121826, | |
| "learning_rate": 7.5e-06, | |
| "loss": 7.8909, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.103092783505154, | |
| "grad_norm": 3.682654619216919, | |
| "learning_rate": 6.875e-06, | |
| "loss": 7.8228, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 6.268041237113402, | |
| "grad_norm": 3.4344913959503174, | |
| "learning_rate": 6.25e-06, | |
| "loss": 7.8066, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.43298969072165, | |
| "grad_norm": 3.7810468673706055, | |
| "learning_rate": 5.625e-06, | |
| "loss": 7.8461, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 6.597938144329897, | |
| "grad_norm": 3.7566187381744385, | |
| "learning_rate": 4.9999999999999996e-06, | |
| "loss": 7.8968, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.762886597938144, | |
| "grad_norm": 3.617682695388794, | |
| "learning_rate": 4.3750000000000005e-06, | |
| "loss": 7.9953, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 6.927835051546392, | |
| "grad_norm": 3.5307486057281494, | |
| "learning_rate": 3.75e-06, | |
| "loss": 7.8764, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.092783505154639, | |
| "grad_norm": 3.624812364578247, | |
| "learning_rate": 3.125e-06, | |
| "loss": 7.7608, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 7.257731958762887, | |
| "grad_norm": 3.573174476623535, | |
| "learning_rate": 2.4999999999999998e-06, | |
| "loss": 7.705, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 7.422680412371134, | |
| "grad_norm": 3.3969686031341553, | |
| "learning_rate": 1.875e-06, | |
| "loss": 7.9902, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 7.587628865979381, | |
| "grad_norm": 3.5248804092407227, | |
| "learning_rate": 1.2499999999999999e-06, | |
| "loss": 7.8566, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 7.752577319587629, | |
| "grad_norm": 3.8090217113494873, | |
| "learning_rate": 6.249999999999999e-07, | |
| "loss": 7.5691, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 7.917525773195877, | |
| "grad_norm": 3.4885289669036865, | |
| "learning_rate": 0.0, | |
| "loss": 7.8552, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 7.917525773195877, | |
| "step": 48, | |
| "total_flos": 962957869218000.0, | |
| "train_loss": 8.716416617234549, | |
| "train_runtime": 7439.5895, | |
| "train_samples_per_second": 0.104, | |
| "train_steps_per_second": 0.006 | |
| } | |
| ], | |
| "logging_steps": 1.0, | |
| "max_steps": 48, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 8, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": false, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 962957869218000.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |