diogopaes10's picture
End of training
dad6435
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 15.0,
"global_step": 375,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 1.9946666666666667e-05,
"loss": 2.3122,
"step": 1
},
{
"epoch": 0.76,
"learning_rate": 1.898666666666667e-05,
"loss": 2.3022,
"step": 19
},
{
"epoch": 0.76,
"eval_accuracy": 0.1,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 33.75697326660156,
"eval_f1": 0.018181818181818184,
"eval_gpu_ram_allocated": 4.172749996185303,
"eval_gpu_ram_cached": 26.759765625,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 45,
"eval_loss": 2.3011605739593506,
"eval_precision": 0.01,
"eval_recall": 0.1,
"eval_runtime": 0.3917,
"eval_samples_per_second": 510.562,
"eval_steps_per_second": 17.87,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 4.445606231689453,
"step": 19
},
{
"epoch": 1.52,
"learning_rate": 1.7973333333333333e-05,
"loss": 2.2979,
"step": 38
},
{
"epoch": 1.52,
"eval_accuracy": 0.155,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 38.592166900634766,
"eval_f1": 0.06350931677018633,
"eval_gpu_ram_allocated": 4.172748565673828,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 43,
"eval_loss": 2.2853753566741943,
"eval_precision": 0.04491869918699187,
"eval_recall": 0.155,
"eval_runtime": 0.3679,
"eval_samples_per_second": 543.585,
"eval_steps_per_second": 19.025,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.034675598144531,
"step": 38
},
{
"epoch": 2.28,
"learning_rate": 1.696e-05,
"loss": 2.2316,
"step": 57
},
{
"epoch": 2.28,
"eval_accuracy": 0.305,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66392517089844,
"eval_f1": 0.2284533029124758,
"eval_gpu_ram_allocated": 4.172748565673828,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 44,
"eval_loss": 2.1098108291625977,
"eval_precision": 0.2806426799007444,
"eval_recall": 0.305,
"eval_runtime": 0.3885,
"eval_samples_per_second": 514.84,
"eval_steps_per_second": 18.019,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.178070068359375,
"step": 57
},
{
"epoch": 3.04,
"learning_rate": 1.5946666666666668e-05,
"loss": 1.9915,
"step": 76
},
{
"epoch": 3.04,
"eval_accuracy": 0.43,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.663944244384766,
"eval_f1": 0.4147692106558386,
"eval_gpu_ram_allocated": 4.172744274139404,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 50,
"eval_loss": 1.8477184772491455,
"eval_precision": 0.5039625709645932,
"eval_recall": 0.43,
"eval_runtime": 0.3674,
"eval_samples_per_second": 544.402,
"eval_steps_per_second": 19.054,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.174091339111328,
"step": 76
},
{
"epoch": 3.8,
"learning_rate": 1.4933333333333335e-05,
"loss": 1.684,
"step": 95
},
{
"epoch": 3.8,
"eval_accuracy": 0.55,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.663944244384766,
"eval_f1": 0.5271704828853498,
"eval_gpu_ram_allocated": 4.172764301300049,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 47,
"eval_loss": 1.6027369499206543,
"eval_precision": 0.5665937491057825,
"eval_recall": 0.55,
"eval_runtime": 0.3642,
"eval_samples_per_second": 549.083,
"eval_steps_per_second": 19.218,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.1766204833984375,
"step": 95
},
{
"epoch": 4.56,
"learning_rate": 1.392e-05,
"loss": 1.3911,
"step": 114
},
{
"epoch": 4.56,
"eval_accuracy": 0.615,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66395950317383,
"eval_f1": 0.6060242411805659,
"eval_gpu_ram_allocated": 4.172757148742676,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 49,
"eval_loss": 1.4364641904830933,
"eval_precision": 0.619891761052247,
"eval_recall": 0.615,
"eval_runtime": 0.3684,
"eval_samples_per_second": 542.848,
"eval_steps_per_second": 19.0,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.174568176269531,
"step": 114
},
{
"epoch": 5.32,
"learning_rate": 1.2906666666666668e-05,
"loss": 1.1477,
"step": 133
},
{
"epoch": 5.32,
"eval_accuracy": 0.615,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.663963317871094,
"eval_f1": 0.6215457086252978,
"eval_gpu_ram_allocated": 4.172830104827881,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 52,
"eval_loss": 1.2565349340438843,
"eval_precision": 0.6418622536733991,
"eval_recall": 0.615,
"eval_runtime": 0.3792,
"eval_samples_per_second": 527.473,
"eval_steps_per_second": 18.462,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.1585845947265625,
"step": 133
},
{
"epoch": 6.08,
"learning_rate": 1.1893333333333335e-05,
"loss": 0.9198,
"step": 152
},
{
"epoch": 6.08,
"eval_accuracy": 0.64,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66396713256836,
"eval_f1": 0.6399837944918036,
"eval_gpu_ram_allocated": 4.172749996185303,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 55,
"eval_loss": 1.175949215888977,
"eval_precision": 0.6532184905737537,
"eval_recall": 0.64,
"eval_runtime": 0.3814,
"eval_samples_per_second": 524.426,
"eval_steps_per_second": 18.355,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.180980682373047,
"step": 152
},
{
"epoch": 6.84,
"learning_rate": 1.0880000000000001e-05,
"loss": 0.7605,
"step": 171
},
{
"epoch": 6.84,
"eval_accuracy": 0.645,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66401672363281,
"eval_f1": 0.6418225015596317,
"eval_gpu_ram_allocated": 4.1727471351623535,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 45,
"eval_loss": 1.112830638885498,
"eval_precision": 0.6564146498179483,
"eval_recall": 0.645,
"eval_runtime": 0.3808,
"eval_samples_per_second": 525.141,
"eval_steps_per_second": 18.38,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.141529083251953,
"step": 171
},
{
"epoch": 7.6,
"learning_rate": 9.866666666666668e-06,
"loss": 0.6093,
"step": 190
},
{
"epoch": 7.6,
"eval_accuracy": 0.67,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66403579711914,
"eval_f1": 0.6677676797410276,
"eval_gpu_ram_allocated": 4.172770023345947,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 43,
"eval_loss": 1.0766719579696655,
"eval_precision": 0.6757531007004691,
"eval_recall": 0.67,
"eval_runtime": 0.383,
"eval_samples_per_second": 522.188,
"eval_steps_per_second": 18.277,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.134746551513672,
"step": 190
},
{
"epoch": 8.36,
"learning_rate": 8.853333333333334e-06,
"loss": 0.5111,
"step": 209
},
{
"epoch": 8.36,
"eval_accuracy": 0.655,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.6640510559082,
"eval_f1": 0.655230999157746,
"eval_gpu_ram_allocated": 4.1727728843688965,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 52,
"eval_loss": 1.1033188104629517,
"eval_precision": 0.6741666516629499,
"eval_recall": 0.655,
"eval_runtime": 0.369,
"eval_samples_per_second": 541.937,
"eval_steps_per_second": 18.968,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.120639801025391,
"step": 209
},
{
"epoch": 9.12,
"learning_rate": 7.840000000000001e-06,
"loss": 0.3828,
"step": 228
},
{
"epoch": 9.12,
"eval_accuracy": 0.69,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.664058685302734,
"eval_f1": 0.6874757485635422,
"eval_gpu_ram_allocated": 4.17274284362793,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 44,
"eval_loss": 1.1062885522842407,
"eval_precision": 0.692681623931624,
"eval_recall": 0.69,
"eval_runtime": 0.373,
"eval_samples_per_second": 536.14,
"eval_steps_per_second": 18.765,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.148380279541016,
"step": 228
},
{
"epoch": 9.88,
"learning_rate": 6.826666666666667e-06,
"loss": 0.3082,
"step": 247
},
{
"epoch": 9.88,
"eval_accuracy": 0.665,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.664058685302734,
"eval_f1": 0.657310556170902,
"eval_gpu_ram_allocated": 4.172765731811523,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 45,
"eval_loss": 1.1239553689956665,
"eval_precision": 0.6595053580899589,
"eval_recall": 0.665,
"eval_runtime": 0.3747,
"eval_samples_per_second": 533.805,
"eval_steps_per_second": 18.683,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.143707275390625,
"step": 247
},
{
"epoch": 10.64,
"learning_rate": 5.813333333333334e-06,
"loss": 0.2716,
"step": 266
},
{
"epoch": 10.64,
"eval_accuracy": 0.665,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66410827636719,
"eval_f1": 0.6603758535972988,
"eval_gpu_ram_allocated": 4.172775745391846,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 45,
"eval_loss": 1.157199501991272,
"eval_precision": 0.666485053212742,
"eval_recall": 0.665,
"eval_runtime": 0.3689,
"eval_samples_per_second": 542.088,
"eval_steps_per_second": 18.973,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.068927764892578,
"step": 266
},
{
"epoch": 11.4,
"learning_rate": 4.800000000000001e-06,
"loss": 0.2442,
"step": 285
},
{
"epoch": 11.4,
"eval_accuracy": 0.675,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66411209106445,
"eval_f1": 0.6764825046795931,
"eval_gpu_ram_allocated": 4.17277717590332,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 42,
"eval_loss": 1.1057575941085815,
"eval_precision": 0.6826701222753855,
"eval_recall": 0.675,
"eval_runtime": 0.3756,
"eval_samples_per_second": 532.455,
"eval_steps_per_second": 18.636,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.031635284423828,
"step": 285
},
{
"epoch": 12.16,
"learning_rate": 3.7866666666666667e-06,
"loss": 0.1791,
"step": 304
},
{
"epoch": 12.16,
"eval_accuracy": 0.645,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.664119720458984,
"eval_f1": 0.6445427433817439,
"eval_gpu_ram_allocated": 4.172751426696777,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 46,
"eval_loss": 1.1455398797988892,
"eval_precision": 0.6514674297225784,
"eval_recall": 0.645,
"eval_runtime": 0.37,
"eval_samples_per_second": 540.499,
"eval_steps_per_second": 18.917,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.071483612060547,
"step": 304
},
{
"epoch": 12.92,
"learning_rate": 2.7733333333333336e-06,
"loss": 0.1604,
"step": 323
},
{
"epoch": 12.92,
"eval_accuracy": 0.66,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66413116455078,
"eval_f1": 0.6578390320866488,
"eval_gpu_ram_allocated": 4.172774314880371,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 57,
"eval_loss": 1.1514325141906738,
"eval_precision": 0.6686316887895163,
"eval_recall": 0.66,
"eval_runtime": 0.368,
"eval_samples_per_second": 543.448,
"eval_steps_per_second": 19.021,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.0727691650390625,
"step": 323
},
{
"epoch": 13.68,
"learning_rate": 1.76e-06,
"loss": 0.1389,
"step": 342
},
{
"epoch": 13.68,
"eval_accuracy": 0.675,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.66417694091797,
"eval_f1": 0.6714518778998741,
"eval_gpu_ram_allocated": 4.172738552093506,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 48,
"eval_loss": 1.160007357597351,
"eval_precision": 0.6807936997642879,
"eval_recall": 0.675,
"eval_runtime": 0.3735,
"eval_samples_per_second": 535.495,
"eval_steps_per_second": 18.742,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.065456390380859,
"step": 342
},
{
"epoch": 14.44,
"learning_rate": 7.466666666666668e-07,
"loss": 0.151,
"step": 361
},
{
"epoch": 14.44,
"eval_accuracy": 0.665,
"eval_disk_space_total": 78.1898422241211,
"eval_disk_space_used": 40.664188385009766,
"eval_f1": 0.6625899013665082,
"eval_gpu_ram_allocated": 4.172741413116455,
"eval_gpu_ram_cached": 26.771484375,
"eval_gpu_ram_total": 39.56402587890625,
"eval_gpu_utilization": 48,
"eval_loss": 1.1573272943496704,
"eval_precision": 0.6686724234321488,
"eval_recall": 0.665,
"eval_runtime": 0.3721,
"eval_samples_per_second": 537.558,
"eval_steps_per_second": 18.815,
"eval_system_ram_total": 83.48074722290039,
"eval_system_ram_used": 5.058799743652344,
"step": 361
}
],
"max_steps": 375,
"num_train_epochs": 15,
"total_flos": 825897508635264.0,
"trial_name": null,
"trial_params": null
}