2942 lines
69 KiB
JSON
2942 lines
69 KiB
JSON
|
{
|
||
|
|
"best_global_step": null,
|
||
|
|
"best_metric": null,
|
||
|
|
"best_model_checkpoint": null,
|
||
|
|
"epoch": 1.0,
|
||
|
|
"eval_steps": 10,
|
||
|
|
"global_step": 369,
|
||
|
|
"is_hyper_param_search": false,
|
||
|
|
"is_local_process_zero": true,
|
||
|
|
"is_world_process_zero": true,
|
||
|
|
"log_history": [
|
||
|
|
{
|
||
|
|
"epoch": 0.002711864406779661,
|
||
|
|
"grad_norm": 2.28125,
|
||
|
|
"learning_rate": 0.0,
|
||
|
|
"loss": 0.7736,
|
||
|
|
"step": 1
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.005423728813559322,
|
||
|
|
"grad_norm": 3.078125,
|
||
|
|
"learning_rate": 2.0000000000000003e-06,
|
||
|
|
"loss": 1.2456,
|
||
|
|
"step": 2
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.008135593220338983,
|
||
|
|
"grad_norm": 2.71875,
|
||
|
|
"learning_rate": 4.000000000000001e-06,
|
||
|
|
"loss": 1.0688,
|
||
|
|
"step": 3
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.010847457627118645,
|
||
|
|
"grad_norm": 2.53125,
|
||
|
|
"learning_rate": 6e-06,
|
||
|
|
"loss": 0.9478,
|
||
|
|
"step": 4
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.013559322033898305,
|
||
|
|
"grad_norm": 2.3125,
|
||
|
|
"learning_rate": 8.000000000000001e-06,
|
||
|
|
"loss": 0.8185,
|
||
|
|
"step": 5
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.016271186440677966,
|
||
|
|
"grad_norm": 2.921875,
|
||
|
|
"learning_rate": 1e-05,
|
||
|
|
"loss": 0.9357,
|
||
|
|
"step": 6
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.018983050847457626,
|
||
|
|
"grad_norm": 2.453125,
|
||
|
|
"learning_rate": 9.972527472527474e-06,
|
||
|
|
"loss": 0.9244,
|
||
|
|
"step": 7
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02169491525423729,
|
||
|
|
"grad_norm": 1.7578125,
|
||
|
|
"learning_rate": 9.945054945054946e-06,
|
||
|
|
"loss": 0.6396,
|
||
|
|
"step": 8
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02440677966101695,
|
||
|
|
"grad_norm": 1.421875,
|
||
|
|
"learning_rate": 9.917582417582419e-06,
|
||
|
|
"loss": 0.6322,
|
||
|
|
"step": 9
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02711864406779661,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 9.890109890109892e-06,
|
||
|
|
"loss": 0.5287,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.02711864406779661,
|
||
|
|
"eval_loss": 0.7306801080703735,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.2959,
|
||
|
|
"eval_samples_per_second": 8.853,
|
||
|
|
"eval_steps_per_second": 4.426,
|
||
|
|
"step": 10
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.029830508474576273,
|
||
|
|
"grad_norm": 1.2578125,
|
||
|
|
"learning_rate": 9.862637362637364e-06,
|
||
|
|
"loss": 0.464,
|
||
|
|
"step": 11
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.03254237288135593,
|
||
|
|
"grad_norm": 1.109375,
|
||
|
|
"learning_rate": 9.835164835164835e-06,
|
||
|
|
"loss": 0.476,
|
||
|
|
"step": 12
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.03525423728813559,
|
||
|
|
"grad_norm": 1.1953125,
|
||
|
|
"learning_rate": 9.807692307692308e-06,
|
||
|
|
"loss": 0.7237,
|
||
|
|
"step": 13
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.03796610169491525,
|
||
|
|
"grad_norm": 1.1484375,
|
||
|
|
"learning_rate": 9.780219780219781e-06,
|
||
|
|
"loss": 0.5612,
|
||
|
|
"step": 14
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.04067796610169491,
|
||
|
|
"grad_norm": 1.046875,
|
||
|
|
"learning_rate": 9.752747252747253e-06,
|
||
|
|
"loss": 0.5471,
|
||
|
|
"step": 15
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.04338983050847458,
|
||
|
|
"grad_norm": 1.03125,
|
||
|
|
"learning_rate": 9.725274725274726e-06,
|
||
|
|
"loss": 0.5104,
|
||
|
|
"step": 16
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.04610169491525424,
|
||
|
|
"grad_norm": 0.90234375,
|
||
|
|
"learning_rate": 9.697802197802198e-06,
|
||
|
|
"loss": 0.4911,
|
||
|
|
"step": 17
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0488135593220339,
|
||
|
|
"grad_norm": 0.82421875,
|
||
|
|
"learning_rate": 9.670329670329671e-06,
|
||
|
|
"loss": 0.3772,
|
||
|
|
"step": 18
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.05152542372881356,
|
||
|
|
"grad_norm": 0.96484375,
|
||
|
|
"learning_rate": 9.642857142857144e-06,
|
||
|
|
"loss": 0.495,
|
||
|
|
"step": 19
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.05423728813559322,
|
||
|
|
"grad_norm": 0.8359375,
|
||
|
|
"learning_rate": 9.615384615384616e-06,
|
||
|
|
"loss": 0.4638,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.05423728813559322,
|
||
|
|
"eval_loss": 0.4768664240837097,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3113,
|
||
|
|
"eval_samples_per_second": 8.841,
|
||
|
|
"eval_steps_per_second": 4.42,
|
||
|
|
"step": 20
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.05694915254237288,
|
||
|
|
"grad_norm": 0.87890625,
|
||
|
|
"learning_rate": 9.587912087912089e-06,
|
||
|
|
"loss": 0.452,
|
||
|
|
"step": 21
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.059661016949152545,
|
||
|
|
"grad_norm": 0.79296875,
|
||
|
|
"learning_rate": 9.560439560439562e-06,
|
||
|
|
"loss": 0.4108,
|
||
|
|
"step": 22
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.062372881355932205,
|
||
|
|
"grad_norm": 0.73828125,
|
||
|
|
"learning_rate": 9.532967032967034e-06,
|
||
|
|
"loss": 0.3777,
|
||
|
|
"step": 23
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.06508474576271187,
|
||
|
|
"grad_norm": 0.8828125,
|
||
|
|
"learning_rate": 9.505494505494505e-06,
|
||
|
|
"loss": 0.577,
|
||
|
|
"step": 24
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.06779661016949153,
|
||
|
|
"grad_norm": 0.73046875,
|
||
|
|
"learning_rate": 9.478021978021978e-06,
|
||
|
|
"loss": 0.4745,
|
||
|
|
"step": 25
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.07050847457627119,
|
||
|
|
"grad_norm": 0.4609375,
|
||
|
|
"learning_rate": 9.450549450549452e-06,
|
||
|
|
"loss": 0.2241,
|
||
|
|
"step": 26
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.07322033898305084,
|
||
|
|
"grad_norm": 0.65625,
|
||
|
|
"learning_rate": 9.423076923076923e-06,
|
||
|
|
"loss": 0.4402,
|
||
|
|
"step": 27
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0759322033898305,
|
||
|
|
"grad_norm": 0.578125,
|
||
|
|
"learning_rate": 9.395604395604396e-06,
|
||
|
|
"loss": 0.4457,
|
||
|
|
"step": 28
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.07864406779661016,
|
||
|
|
"grad_norm": 0.482421875,
|
||
|
|
"learning_rate": 9.36813186813187e-06,
|
||
|
|
"loss": 0.2633,
|
||
|
|
"step": 29
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.08135593220338982,
|
||
|
|
"grad_norm": 0.55859375,
|
||
|
|
"learning_rate": 9.340659340659341e-06,
|
||
|
|
"loss": 0.3224,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.08135593220338982,
|
||
|
|
"eval_loss": 0.3509748876094818,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.2986,
|
||
|
|
"eval_samples_per_second": 8.851,
|
||
|
|
"eval_steps_per_second": 4.425,
|
||
|
|
"step": 30
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0840677966101695,
|
||
|
|
"grad_norm": 0.60546875,
|
||
|
|
"learning_rate": 9.313186813186814e-06,
|
||
|
|
"loss": 0.442,
|
||
|
|
"step": 31
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.08677966101694916,
|
||
|
|
"grad_norm": 0.515625,
|
||
|
|
"learning_rate": 9.285714285714288e-06,
|
||
|
|
"loss": 0.3376,
|
||
|
|
"step": 32
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.08949152542372882,
|
||
|
|
"grad_norm": 0.48828125,
|
||
|
|
"learning_rate": 9.258241758241759e-06,
|
||
|
|
"loss": 0.3107,
|
||
|
|
"step": 33
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.09220338983050848,
|
||
|
|
"grad_norm": 0.41015625,
|
||
|
|
"learning_rate": 9.230769230769232e-06,
|
||
|
|
"loss": 0.2595,
|
||
|
|
"step": 34
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.09491525423728814,
|
||
|
|
"grad_norm": 0.515625,
|
||
|
|
"learning_rate": 9.203296703296704e-06,
|
||
|
|
"loss": 0.3329,
|
||
|
|
"step": 35
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.0976271186440678,
|
||
|
|
"grad_norm": 0.4453125,
|
||
|
|
"learning_rate": 9.175824175824175e-06,
|
||
|
|
"loss": 0.3234,
|
||
|
|
"step": 36
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10033898305084746,
|
||
|
|
"grad_norm": 0.466796875,
|
||
|
|
"learning_rate": 9.148351648351649e-06,
|
||
|
|
"loss": 0.3335,
|
||
|
|
"step": 37
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10305084745762712,
|
||
|
|
"grad_norm": 0.31640625,
|
||
|
|
"learning_rate": 9.120879120879122e-06,
|
||
|
|
"loss": 0.1704,
|
||
|
|
"step": 38
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10576271186440678,
|
||
|
|
"grad_norm": 0.34765625,
|
||
|
|
"learning_rate": 9.093406593406593e-06,
|
||
|
|
"loss": 0.2484,
|
||
|
|
"step": 39
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10847457627118644,
|
||
|
|
"grad_norm": 0.53515625,
|
||
|
|
"learning_rate": 9.065934065934067e-06,
|
||
|
|
"loss": 0.4135,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.10847457627118644,
|
||
|
|
"eval_loss": 0.30060264468193054,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.2934,
|
||
|
|
"eval_samples_per_second": 8.855,
|
||
|
|
"eval_steps_per_second": 4.427,
|
||
|
|
"step": 40
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1111864406779661,
|
||
|
|
"grad_norm": 0.333984375,
|
||
|
|
"learning_rate": 9.03846153846154e-06,
|
||
|
|
"loss": 0.2265,
|
||
|
|
"step": 41
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.11389830508474576,
|
||
|
|
"grad_norm": 0.421875,
|
||
|
|
"learning_rate": 9.010989010989011e-06,
|
||
|
|
"loss": 0.3255,
|
||
|
|
"step": 42
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.11661016949152542,
|
||
|
|
"grad_norm": 0.4453125,
|
||
|
|
"learning_rate": 8.983516483516484e-06,
|
||
|
|
"loss": 0.3604,
|
||
|
|
"step": 43
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.11932203389830509,
|
||
|
|
"grad_norm": 0.423828125,
|
||
|
|
"learning_rate": 8.956043956043958e-06,
|
||
|
|
"loss": 0.3244,
|
||
|
|
"step": 44
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.12203389830508475,
|
||
|
|
"grad_norm": 0.431640625,
|
||
|
|
"learning_rate": 8.92857142857143e-06,
|
||
|
|
"loss": 0.2405,
|
||
|
|
"step": 45
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.12474576271186441,
|
||
|
|
"grad_norm": 0.365234375,
|
||
|
|
"learning_rate": 8.9010989010989e-06,
|
||
|
|
"loss": 0.3178,
|
||
|
|
"step": 46
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.12745762711864406,
|
||
|
|
"grad_norm": 0.609375,
|
||
|
|
"learning_rate": 8.873626373626374e-06,
|
||
|
|
"loss": 0.5563,
|
||
|
|
"step": 47
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13016949152542373,
|
||
|
|
"grad_norm": 0.337890625,
|
||
|
|
"learning_rate": 8.846153846153847e-06,
|
||
|
|
"loss": 0.1937,
|
||
|
|
"step": 48
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13288135593220338,
|
||
|
|
"grad_norm": 0.455078125,
|
||
|
|
"learning_rate": 8.818681318681319e-06,
|
||
|
|
"loss": 0.3613,
|
||
|
|
"step": 49
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13559322033898305,
|
||
|
|
"grad_norm": 0.326171875,
|
||
|
|
"learning_rate": 8.791208791208792e-06,
|
||
|
|
"loss": 0.1717,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13559322033898305,
|
||
|
|
"eval_loss": 0.27889174222946167,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.2964,
|
||
|
|
"eval_samples_per_second": 8.852,
|
||
|
|
"eval_steps_per_second": 4.426,
|
||
|
|
"step": 50
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.13830508474576272,
|
||
|
|
"grad_norm": 0.296875,
|
||
|
|
"learning_rate": 8.763736263736265e-06,
|
||
|
|
"loss": 0.1388,
|
||
|
|
"step": 51
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.14101694915254237,
|
||
|
|
"grad_norm": 0.369140625,
|
||
|
|
"learning_rate": 8.736263736263737e-06,
|
||
|
|
"loss": 0.3083,
|
||
|
|
"step": 52
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.14372881355932204,
|
||
|
|
"grad_norm": 0.447265625,
|
||
|
|
"learning_rate": 8.70879120879121e-06,
|
||
|
|
"loss": 0.3756,
|
||
|
|
"step": 53
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1464406779661017,
|
||
|
|
"grad_norm": 0.298828125,
|
||
|
|
"learning_rate": 8.681318681318681e-06,
|
||
|
|
"loss": 0.279,
|
||
|
|
"step": 54
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.14915254237288136,
|
||
|
|
"grad_norm": 0.28125,
|
||
|
|
"learning_rate": 8.653846153846155e-06,
|
||
|
|
"loss": 0.1603,
|
||
|
|
"step": 55
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.151864406779661,
|
||
|
|
"grad_norm": 0.333984375,
|
||
|
|
"learning_rate": 8.626373626373628e-06,
|
||
|
|
"loss": 0.2036,
|
||
|
|
"step": 56
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.15457627118644068,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 8.5989010989011e-06,
|
||
|
|
"loss": 0.304,
|
||
|
|
"step": 57
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.15728813559322033,
|
||
|
|
"grad_norm": 0.283203125,
|
||
|
|
"learning_rate": 8.571428571428571e-06,
|
||
|
|
"loss": 0.164,
|
||
|
|
"step": 58
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16,
|
||
|
|
"grad_norm": 0.388671875,
|
||
|
|
"learning_rate": 8.543956043956044e-06,
|
||
|
|
"loss": 0.2686,
|
||
|
|
"step": 59
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16271186440677965,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 8.516483516483517e-06,
|
||
|
|
"loss": 0.2263,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16271186440677965,
|
||
|
|
"eval_loss": 0.26370447874069214,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.4114,
|
||
|
|
"eval_samples_per_second": 8.763,
|
||
|
|
"eval_steps_per_second": 4.382,
|
||
|
|
"step": 60
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.16542372881355932,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 8.489010989010989e-06,
|
||
|
|
"loss": 0.2605,
|
||
|
|
"step": 61
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.168135593220339,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 8.461538461538462e-06,
|
||
|
|
"loss": 0.2453,
|
||
|
|
"step": 62
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.17084745762711864,
|
||
|
|
"grad_norm": 0.29296875,
|
||
|
|
"learning_rate": 8.434065934065935e-06,
|
||
|
|
"loss": 0.1774,
|
||
|
|
"step": 63
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.17355932203389832,
|
||
|
|
"grad_norm": 0.314453125,
|
||
|
|
"learning_rate": 8.406593406593407e-06,
|
||
|
|
"loss": 0.2007,
|
||
|
|
"step": 64
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.17627118644067796,
|
||
|
|
"grad_norm": 0.255859375,
|
||
|
|
"learning_rate": 8.37912087912088e-06,
|
||
|
|
"loss": 0.1479,
|
||
|
|
"step": 65
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.17898305084745764,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 8.351648351648353e-06,
|
||
|
|
"loss": 0.1971,
|
||
|
|
"step": 66
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18169491525423728,
|
||
|
|
"grad_norm": 0.25390625,
|
||
|
|
"learning_rate": 8.324175824175825e-06,
|
||
|
|
"loss": 0.1411,
|
||
|
|
"step": 67
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18440677966101696,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 8.296703296703298e-06,
|
||
|
|
"loss": 0.2188,
|
||
|
|
"step": 68
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1871186440677966,
|
||
|
|
"grad_norm": 0.287109375,
|
||
|
|
"learning_rate": 8.26923076923077e-06,
|
||
|
|
"loss": 0.157,
|
||
|
|
"step": 69
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18983050847457628,
|
||
|
|
"grad_norm": 0.39453125,
|
||
|
|
"learning_rate": 8.241758241758243e-06,
|
||
|
|
"loss": 0.3359,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.18983050847457628,
|
||
|
|
"eval_loss": 0.25340646505355835,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3213,
|
||
|
|
"eval_samples_per_second": 8.833,
|
||
|
|
"eval_steps_per_second": 4.416,
|
||
|
|
"step": 70
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.19254237288135592,
|
||
|
|
"grad_norm": 0.38671875,
|
||
|
|
"learning_rate": 8.214285714285714e-06,
|
||
|
|
"loss": 0.2741,
|
||
|
|
"step": 71
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.1952542372881356,
|
||
|
|
"grad_norm": 0.373046875,
|
||
|
|
"learning_rate": 8.186813186813188e-06,
|
||
|
|
"loss": 0.2986,
|
||
|
|
"step": 72
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.19796610169491524,
|
||
|
|
"grad_norm": 0.345703125,
|
||
|
|
"learning_rate": 8.15934065934066e-06,
|
||
|
|
"loss": 0.2458,
|
||
|
|
"step": 73
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.20067796610169492,
|
||
|
|
"grad_norm": 0.31640625,
|
||
|
|
"learning_rate": 8.131868131868132e-06,
|
||
|
|
"loss": 0.2598,
|
||
|
|
"step": 74
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2033898305084746,
|
||
|
|
"grad_norm": 0.267578125,
|
||
|
|
"learning_rate": 8.104395604395605e-06,
|
||
|
|
"loss": 0.1463,
|
||
|
|
"step": 75
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.20610169491525424,
|
||
|
|
"grad_norm": 0.287109375,
|
||
|
|
"learning_rate": 8.076923076923077e-06,
|
||
|
|
"loss": 0.166,
|
||
|
|
"step": 76
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2088135593220339,
|
||
|
|
"grad_norm": 0.4140625,
|
||
|
|
"learning_rate": 8.04945054945055e-06,
|
||
|
|
"loss": 0.2918,
|
||
|
|
"step": 77
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21152542372881356,
|
||
|
|
"grad_norm": 0.296875,
|
||
|
|
"learning_rate": 8.021978021978023e-06,
|
||
|
|
"loss": 0.2058,
|
||
|
|
"step": 78
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21423728813559323,
|
||
|
|
"grad_norm": 0.318359375,
|
||
|
|
"learning_rate": 7.994505494505495e-06,
|
||
|
|
"loss": 0.1957,
|
||
|
|
"step": 79
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21694915254237288,
|
||
|
|
"grad_norm": 0.34765625,
|
||
|
|
"learning_rate": 7.967032967032966e-06,
|
||
|
|
"loss": 0.2418,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21694915254237288,
|
||
|
|
"eval_loss": 0.2449192851781845,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3151,
|
||
|
|
"eval_samples_per_second": 8.838,
|
||
|
|
"eval_steps_per_second": 4.419,
|
||
|
|
"step": 80
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.21966101694915255,
|
||
|
|
"grad_norm": 0.416015625,
|
||
|
|
"learning_rate": 7.93956043956044e-06,
|
||
|
|
"loss": 0.2543,
|
||
|
|
"step": 81
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2223728813559322,
|
||
|
|
"grad_norm": 0.353515625,
|
||
|
|
"learning_rate": 7.912087912087913e-06,
|
||
|
|
"loss": 0.2527,
|
||
|
|
"step": 82
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.22508474576271187,
|
||
|
|
"grad_norm": 0.400390625,
|
||
|
|
"learning_rate": 7.884615384615384e-06,
|
||
|
|
"loss": 0.23,
|
||
|
|
"step": 83
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.22779661016949151,
|
||
|
|
"grad_norm": 0.447265625,
|
||
|
|
"learning_rate": 7.857142857142858e-06,
|
||
|
|
"loss": 0.2714,
|
||
|
|
"step": 84
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2305084745762712,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 7.829670329670331e-06,
|
||
|
|
"loss": 0.2177,
|
||
|
|
"step": 85
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.23322033898305083,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 7.802197802197802e-06,
|
||
|
|
"loss": 0.2209,
|
||
|
|
"step": 86
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2359322033898305,
|
||
|
|
"grad_norm": 0.4140625,
|
||
|
|
"learning_rate": 7.774725274725276e-06,
|
||
|
|
"loss": 0.2322,
|
||
|
|
"step": 87
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.23864406779661018,
|
||
|
|
"grad_norm": 0.396484375,
|
||
|
|
"learning_rate": 7.747252747252749e-06,
|
||
|
|
"loss": 0.2553,
|
||
|
|
"step": 88
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.24135593220338983,
|
||
|
|
"grad_norm": 0.35546875,
|
||
|
|
"learning_rate": 7.71978021978022e-06,
|
||
|
|
"loss": 0.2081,
|
||
|
|
"step": 89
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2440677966101695,
|
||
|
|
"grad_norm": 0.34375,
|
||
|
|
"learning_rate": 7.692307692307694e-06,
|
||
|
|
"loss": 0.159,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2440677966101695,
|
||
|
|
"eval_loss": 0.23829074203968048,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3009,
|
||
|
|
"eval_samples_per_second": 8.849,
|
||
|
|
"eval_steps_per_second": 4.424,
|
||
|
|
"step": 90
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.24677966101694915,
|
||
|
|
"grad_norm": 0.375,
|
||
|
|
"learning_rate": 7.664835164835167e-06,
|
||
|
|
"loss": 0.1883,
|
||
|
|
"step": 91
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.24949152542372882,
|
||
|
|
"grad_norm": 0.330078125,
|
||
|
|
"learning_rate": 7.637362637362638e-06,
|
||
|
|
"loss": 0.1843,
|
||
|
|
"step": 92
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2522033898305085,
|
||
|
|
"grad_norm": 0.42578125,
|
||
|
|
"learning_rate": 7.60989010989011e-06,
|
||
|
|
"loss": 0.2159,
|
||
|
|
"step": 93
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2549152542372881,
|
||
|
|
"grad_norm": 0.447265625,
|
||
|
|
"learning_rate": 7.582417582417583e-06,
|
||
|
|
"loss": 0.2595,
|
||
|
|
"step": 94
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2576271186440678,
|
||
|
|
"grad_norm": 0.37109375,
|
||
|
|
"learning_rate": 7.5549450549450554e-06,
|
||
|
|
"loss": 0.2701,
|
||
|
|
"step": 95
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.26033898305084746,
|
||
|
|
"grad_norm": 0.40234375,
|
||
|
|
"learning_rate": 7.527472527472528e-06,
|
||
|
|
"loss": 0.2664,
|
||
|
|
"step": 96
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.26305084745762713,
|
||
|
|
"grad_norm": 0.404296875,
|
||
|
|
"learning_rate": 7.500000000000001e-06,
|
||
|
|
"loss": 0.2232,
|
||
|
|
"step": 97
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.26576271186440675,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 7.472527472527473e-06,
|
||
|
|
"loss": 0.2196,
|
||
|
|
"step": 98
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2684745762711864,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 7.445054945054946e-06,
|
||
|
|
"loss": 0.2688,
|
||
|
|
"step": 99
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2711864406779661,
|
||
|
|
"grad_norm": 0.283203125,
|
||
|
|
"learning_rate": 7.417582417582418e-06,
|
||
|
|
"loss": 0.1546,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2711864406779661,
|
||
|
|
"eval_loss": 0.2324570119380951,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3447,
|
||
|
|
"eval_samples_per_second": 8.815,
|
||
|
|
"eval_steps_per_second": 4.407,
|
||
|
|
"step": 100
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2738983050847458,
|
||
|
|
"grad_norm": 0.423828125,
|
||
|
|
"learning_rate": 7.390109890109891e-06,
|
||
|
|
"loss": 0.2543,
|
||
|
|
"step": 101
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.27661016949152545,
|
||
|
|
"grad_norm": 0.451171875,
|
||
|
|
"learning_rate": 7.362637362637364e-06,
|
||
|
|
"loss": 0.3211,
|
||
|
|
"step": 102
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.27932203389830507,
|
||
|
|
"grad_norm": 0.36328125,
|
||
|
|
"learning_rate": 7.335164835164835e-06,
|
||
|
|
"loss": 0.2333,
|
||
|
|
"step": 103
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.28203389830508474,
|
||
|
|
"grad_norm": 0.38671875,
|
||
|
|
"learning_rate": 7.307692307692308e-06,
|
||
|
|
"loss": 0.2475,
|
||
|
|
"step": 104
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2847457627118644,
|
||
|
|
"grad_norm": 0.412109375,
|
||
|
|
"learning_rate": 7.280219780219781e-06,
|
||
|
|
"loss": 0.3498,
|
||
|
|
"step": 105
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2874576271186441,
|
||
|
|
"grad_norm": 0.376953125,
|
||
|
|
"learning_rate": 7.252747252747253e-06,
|
||
|
|
"loss": 0.2406,
|
||
|
|
"step": 106
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2901694915254237,
|
||
|
|
"grad_norm": 0.390625,
|
||
|
|
"learning_rate": 7.225274725274726e-06,
|
||
|
|
"loss": 0.2182,
|
||
|
|
"step": 107
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2928813559322034,
|
||
|
|
"grad_norm": 0.3203125,
|
||
|
|
"learning_rate": 7.197802197802198e-06,
|
||
|
|
"loss": 0.1852,
|
||
|
|
"step": 108
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.29559322033898305,
|
||
|
|
"grad_norm": 0.337890625,
|
||
|
|
"learning_rate": 7.170329670329671e-06,
|
||
|
|
"loss": 0.1863,
|
||
|
|
"step": 109
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2983050847457627,
|
||
|
|
"grad_norm": 0.412109375,
|
||
|
|
"learning_rate": 7.1428571428571436e-06,
|
||
|
|
"loss": 0.26,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.2983050847457627,
|
||
|
|
"eval_loss": 0.22855669260025024,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.438,
|
||
|
|
"eval_samples_per_second": 8.743,
|
||
|
|
"eval_steps_per_second": 4.371,
|
||
|
|
"step": 110
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.30101694915254235,
|
||
|
|
"grad_norm": 0.361328125,
|
||
|
|
"learning_rate": 7.115384615384616e-06,
|
||
|
|
"loss": 0.1708,
|
||
|
|
"step": 111
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.303728813559322,
|
||
|
|
"grad_norm": 0.388671875,
|
||
|
|
"learning_rate": 7.087912087912089e-06,
|
||
|
|
"loss": 0.2213,
|
||
|
|
"step": 112
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3064406779661017,
|
||
|
|
"grad_norm": 0.28515625,
|
||
|
|
"learning_rate": 7.0604395604395615e-06,
|
||
|
|
"loss": 0.135,
|
||
|
|
"step": 113
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.30915254237288137,
|
||
|
|
"grad_norm": 0.3046875,
|
||
|
|
"learning_rate": 7.032967032967034e-06,
|
||
|
|
"loss": 0.1405,
|
||
|
|
"step": 114
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.31186440677966104,
|
||
|
|
"grad_norm": 0.431640625,
|
||
|
|
"learning_rate": 7.005494505494505e-06,
|
||
|
|
"loss": 0.2475,
|
||
|
|
"step": 115
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.31457627118644066,
|
||
|
|
"grad_norm": 0.2333984375,
|
||
|
|
"learning_rate": 6.978021978021979e-06,
|
||
|
|
"loss": 0.0942,
|
||
|
|
"step": 116
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.31728813559322033,
|
||
|
|
"grad_norm": 0.318359375,
|
||
|
|
"learning_rate": 6.950549450549451e-06,
|
||
|
|
"loss": 0.1643,
|
||
|
|
"step": 117
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.32,
|
||
|
|
"grad_norm": 0.2255859375,
|
||
|
|
"learning_rate": 6.923076923076923e-06,
|
||
|
|
"loss": 0.0807,
|
||
|
|
"step": 118
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3227118644067797,
|
||
|
|
"grad_norm": 0.462890625,
|
||
|
|
"learning_rate": 6.895604395604396e-06,
|
||
|
|
"loss": 0.2953,
|
||
|
|
"step": 119
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3254237288135593,
|
||
|
|
"grad_norm": 0.31640625,
|
||
|
|
"learning_rate": 6.868131868131869e-06,
|
||
|
|
"loss": 0.1632,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3254237288135593,
|
||
|
|
"eval_loss": 0.22415581345558167,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5405,
|
||
|
|
"eval_samples_per_second": 8.665,
|
||
|
|
"eval_steps_per_second": 4.333,
|
||
|
|
"step": 120
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.328135593220339,
|
||
|
|
"grad_norm": 0.49609375,
|
||
|
|
"learning_rate": 6.840659340659341e-06,
|
||
|
|
"loss": 0.2407,
|
||
|
|
"step": 121
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.33084745762711865,
|
||
|
|
"grad_norm": 0.365234375,
|
||
|
|
"learning_rate": 6.813186813186814e-06,
|
||
|
|
"loss": 0.229,
|
||
|
|
"step": 122
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3335593220338983,
|
||
|
|
"grad_norm": 0.361328125,
|
||
|
|
"learning_rate": 6.785714285714287e-06,
|
||
|
|
"loss": 0.2249,
|
||
|
|
"step": 123
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.336271186440678,
|
||
|
|
"grad_norm": 0.369140625,
|
||
|
|
"learning_rate": 6.758241758241759e-06,
|
||
|
|
"loss": 0.1877,
|
||
|
|
"step": 124
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3389830508474576,
|
||
|
|
"grad_norm": 0.32421875,
|
||
|
|
"learning_rate": 6.730769230769232e-06,
|
||
|
|
"loss": 0.1355,
|
||
|
|
"step": 125
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3416949152542373,
|
||
|
|
"grad_norm": 0.494140625,
|
||
|
|
"learning_rate": 6.703296703296703e-06,
|
||
|
|
"loss": 0.3479,
|
||
|
|
"step": 126
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.34440677966101696,
|
||
|
|
"grad_norm": 0.30859375,
|
||
|
|
"learning_rate": 6.6758241758241756e-06,
|
||
|
|
"loss": 0.2047,
|
||
|
|
"step": 127
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.34711864406779663,
|
||
|
|
"grad_norm": 0.36328125,
|
||
|
|
"learning_rate": 6.648351648351649e-06,
|
||
|
|
"loss": 0.1821,
|
||
|
|
"step": 128
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.34983050847457625,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 6.620879120879121e-06,
|
||
|
|
"loss": 0.232,
|
||
|
|
"step": 129
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3525423728813559,
|
||
|
|
"grad_norm": 0.34765625,
|
||
|
|
"learning_rate": 6.5934065934065935e-06,
|
||
|
|
"loss": 0.2048,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3525423728813559,
|
||
|
|
"eval_loss": 0.2212015688419342,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.7884,
|
||
|
|
"eval_samples_per_second": 8.483,
|
||
|
|
"eval_steps_per_second": 4.241,
|
||
|
|
"step": 130
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3552542372881356,
|
||
|
|
"grad_norm": 0.408203125,
|
||
|
|
"learning_rate": 6.565934065934067e-06,
|
||
|
|
"loss": 0.2823,
|
||
|
|
"step": 131
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3579661016949153,
|
||
|
|
"grad_norm": 0.291015625,
|
||
|
|
"learning_rate": 6.538461538461539e-06,
|
||
|
|
"loss": 0.1487,
|
||
|
|
"step": 132
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3606779661016949,
|
||
|
|
"grad_norm": 0.46484375,
|
||
|
|
"learning_rate": 6.5109890109890115e-06,
|
||
|
|
"loss": 0.2647,
|
||
|
|
"step": 133
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.36338983050847457,
|
||
|
|
"grad_norm": 0.279296875,
|
||
|
|
"learning_rate": 6.483516483516485e-06,
|
||
|
|
"loss": 0.1578,
|
||
|
|
"step": 134
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.36610169491525424,
|
||
|
|
"grad_norm": 0.34765625,
|
||
|
|
"learning_rate": 6.456043956043957e-06,
|
||
|
|
"loss": 0.1861,
|
||
|
|
"step": 135
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3688135593220339,
|
||
|
|
"grad_norm": 0.341796875,
|
||
|
|
"learning_rate": 6.4285714285714295e-06,
|
||
|
|
"loss": 0.2132,
|
||
|
|
"step": 136
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3715254237288136,
|
||
|
|
"grad_norm": 0.384765625,
|
||
|
|
"learning_rate": 6.401098901098901e-06,
|
||
|
|
"loss": 0.2278,
|
||
|
|
"step": 137
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3742372881355932,
|
||
|
|
"grad_norm": 0.302734375,
|
||
|
|
"learning_rate": 6.373626373626373e-06,
|
||
|
|
"loss": 0.1749,
|
||
|
|
"step": 138
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3769491525423729,
|
||
|
|
"grad_norm": 0.40625,
|
||
|
|
"learning_rate": 6.3461538461538466e-06,
|
||
|
|
"loss": 0.2314,
|
||
|
|
"step": 139
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.37966101694915255,
|
||
|
|
"grad_norm": 0.384765625,
|
||
|
|
"learning_rate": 6.318681318681319e-06,
|
||
|
|
"loss": 0.2517,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.37966101694915255,
|
||
|
|
"eval_loss": 0.21904973685741425,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3876,
|
||
|
|
"eval_samples_per_second": 8.781,
|
||
|
|
"eval_steps_per_second": 4.391,
|
||
|
|
"step": 140
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3823728813559322,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 6.291208791208791e-06,
|
||
|
|
"loss": 0.2115,
|
||
|
|
"step": 141
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.38508474576271184,
|
||
|
|
"grad_norm": 0.41796875,
|
||
|
|
"learning_rate": 6.2637362637362645e-06,
|
||
|
|
"loss": 0.1945,
|
||
|
|
"step": 142
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3877966101694915,
|
||
|
|
"grad_norm": 0.392578125,
|
||
|
|
"learning_rate": 6.236263736263737e-06,
|
||
|
|
"loss": 0.239,
|
||
|
|
"step": 143
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3905084745762712,
|
||
|
|
"grad_norm": 0.458984375,
|
||
|
|
"learning_rate": 6.208791208791209e-06,
|
||
|
|
"loss": 0.356,
|
||
|
|
"step": 144
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.39322033898305087,
|
||
|
|
"grad_norm": 0.388671875,
|
||
|
|
"learning_rate": 6.181318681318682e-06,
|
||
|
|
"loss": 0.2406,
|
||
|
|
"step": 145
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.3959322033898305,
|
||
|
|
"grad_norm": 0.28125,
|
||
|
|
"learning_rate": 6.153846153846155e-06,
|
||
|
|
"loss": 0.1312,
|
||
|
|
"step": 146
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.39864406779661016,
|
||
|
|
"grad_norm": 0.38671875,
|
||
|
|
"learning_rate": 6.126373626373627e-06,
|
||
|
|
"loss": 0.2204,
|
||
|
|
"step": 147
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.40135593220338983,
|
||
|
|
"grad_norm": 0.333984375,
|
||
|
|
"learning_rate": 6.0989010989011e-06,
|
||
|
|
"loss": 0.1769,
|
||
|
|
"step": 148
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4040677966101695,
|
||
|
|
"grad_norm": 0.345703125,
|
||
|
|
"learning_rate": 6.071428571428571e-06,
|
||
|
|
"loss": 0.1817,
|
||
|
|
"step": 149
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4067796610169492,
|
||
|
|
"grad_norm": 0.51953125,
|
||
|
|
"learning_rate": 6.043956043956044e-06,
|
||
|
|
"loss": 0.3207,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4067796610169492,
|
||
|
|
"eval_loss": 0.21635417640209198,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3802,
|
||
|
|
"eval_samples_per_second": 8.787,
|
||
|
|
"eval_steps_per_second": 4.394,
|
||
|
|
"step": 150
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4094915254237288,
|
||
|
|
"grad_norm": 0.341796875,
|
||
|
|
"learning_rate": 6.016483516483517e-06,
|
||
|
|
"loss": 0.1799,
|
||
|
|
"step": 151
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.41220338983050847,
|
||
|
|
"grad_norm": 0.484375,
|
||
|
|
"learning_rate": 5.989010989010989e-06,
|
||
|
|
"loss": 0.2839,
|
||
|
|
"step": 152
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.41491525423728814,
|
||
|
|
"grad_norm": 0.255859375,
|
||
|
|
"learning_rate": 5.961538461538462e-06,
|
||
|
|
"loss": 0.1082,
|
||
|
|
"step": 153
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4176271186440678,
|
||
|
|
"grad_norm": 0.369140625,
|
||
|
|
"learning_rate": 5.934065934065935e-06,
|
||
|
|
"loss": 0.2106,
|
||
|
|
"step": 154
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.42033898305084744,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 5.906593406593407e-06,
|
||
|
|
"loss": 0.2032,
|
||
|
|
"step": 155
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4230508474576271,
|
||
|
|
"grad_norm": 0.400390625,
|
||
|
|
"learning_rate": 5.8791208791208794e-06,
|
||
|
|
"loss": 0.2513,
|
||
|
|
"step": 156
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4257627118644068,
|
||
|
|
"grad_norm": 0.265625,
|
||
|
|
"learning_rate": 5.851648351648353e-06,
|
||
|
|
"loss": 0.1126,
|
||
|
|
"step": 157
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.42847457627118646,
|
||
|
|
"grad_norm": 0.408203125,
|
||
|
|
"learning_rate": 5.824175824175825e-06,
|
||
|
|
"loss": 0.2326,
|
||
|
|
"step": 158
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4311864406779661,
|
||
|
|
"grad_norm": 0.341796875,
|
||
|
|
"learning_rate": 5.796703296703297e-06,
|
||
|
|
"loss": 0.161,
|
||
|
|
"step": 159
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.43389830508474575,
|
||
|
|
"grad_norm": 0.373046875,
|
||
|
|
"learning_rate": 5.769230769230769e-06,
|
||
|
|
"loss": 0.2494,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.43389830508474575,
|
||
|
|
"eval_loss": 0.21449893712997437,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3764,
|
||
|
|
"eval_samples_per_second": 8.79,
|
||
|
|
"eval_steps_per_second": 4.395,
|
||
|
|
"step": 160
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4366101694915254,
|
||
|
|
"grad_norm": 0.32421875,
|
||
|
|
"learning_rate": 5.741758241758242e-06,
|
||
|
|
"loss": 0.1967,
|
||
|
|
"step": 161
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4393220338983051,
|
||
|
|
"grad_norm": 0.33984375,
|
||
|
|
"learning_rate": 5.7142857142857145e-06,
|
||
|
|
"loss": 0.172,
|
||
|
|
"step": 162
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.44203389830508477,
|
||
|
|
"grad_norm": 0.3203125,
|
||
|
|
"learning_rate": 5.686813186813187e-06,
|
||
|
|
"loss": 0.1679,
|
||
|
|
"step": 163
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4447457627118644,
|
||
|
|
"grad_norm": 0.451171875,
|
||
|
|
"learning_rate": 5.65934065934066e-06,
|
||
|
|
"loss": 0.2689,
|
||
|
|
"step": 164
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.44745762711864406,
|
||
|
|
"grad_norm": 0.35546875,
|
||
|
|
"learning_rate": 5.6318681318681325e-06,
|
||
|
|
"loss": 0.1948,
|
||
|
|
"step": 165
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.45016949152542374,
|
||
|
|
"grad_norm": 0.46484375,
|
||
|
|
"learning_rate": 5.604395604395605e-06,
|
||
|
|
"loss": 0.2931,
|
||
|
|
"step": 166
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4528813559322034,
|
||
|
|
"grad_norm": 0.408203125,
|
||
|
|
"learning_rate": 5.576923076923077e-06,
|
||
|
|
"loss": 0.1679,
|
||
|
|
"step": 167
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.45559322033898303,
|
||
|
|
"grad_norm": 0.384765625,
|
||
|
|
"learning_rate": 5.5494505494505504e-06,
|
||
|
|
"loss": 0.2063,
|
||
|
|
"step": 168
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4583050847457627,
|
||
|
|
"grad_norm": 0.51171875,
|
||
|
|
"learning_rate": 5.521978021978023e-06,
|
||
|
|
"loss": 0.2895,
|
||
|
|
"step": 169
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4610169491525424,
|
||
|
|
"grad_norm": 0.365234375,
|
||
|
|
"learning_rate": 5.494505494505495e-06,
|
||
|
|
"loss": 0.1781,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4610169491525424,
|
||
|
|
"eval_loss": 0.2126615196466446,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3672,
|
||
|
|
"eval_samples_per_second": 8.797,
|
||
|
|
"eval_steps_per_second": 4.399,
|
||
|
|
"step": 170
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.46372881355932205,
|
||
|
|
"grad_norm": 0.375,
|
||
|
|
"learning_rate": 5.467032967032967e-06,
|
||
|
|
"loss": 0.1829,
|
||
|
|
"step": 171
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.46644067796610167,
|
||
|
|
"grad_norm": 0.396484375,
|
||
|
|
"learning_rate": 5.43956043956044e-06,
|
||
|
|
"loss": 0.2247,
|
||
|
|
"step": 172
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.46915254237288134,
|
||
|
|
"grad_norm": 0.4140625,
|
||
|
|
"learning_rate": 5.412087912087912e-06,
|
||
|
|
"loss": 0.2509,
|
||
|
|
"step": 173
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.471864406779661,
|
||
|
|
"grad_norm": 0.455078125,
|
||
|
|
"learning_rate": 5.384615384615385e-06,
|
||
|
|
"loss": 0.248,
|
||
|
|
"step": 174
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4745762711864407,
|
||
|
|
"grad_norm": 0.439453125,
|
||
|
|
"learning_rate": 5.357142857142857e-06,
|
||
|
|
"loss": 0.2222,
|
||
|
|
"step": 175
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.47728813559322036,
|
||
|
|
"grad_norm": 0.404296875,
|
||
|
|
"learning_rate": 5.32967032967033e-06,
|
||
|
|
"loss": 0.1882,
|
||
|
|
"step": 176
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.48,
|
||
|
|
"grad_norm": 0.3984375,
|
||
|
|
"learning_rate": 5.302197802197803e-06,
|
||
|
|
"loss": 0.1951,
|
||
|
|
"step": 177
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.48271186440677966,
|
||
|
|
"grad_norm": 0.43359375,
|
||
|
|
"learning_rate": 5.274725274725275e-06,
|
||
|
|
"loss": 0.1958,
|
||
|
|
"step": 178
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.48542372881355933,
|
||
|
|
"grad_norm": 0.359375,
|
||
|
|
"learning_rate": 5.247252747252748e-06,
|
||
|
|
"loss": 0.164,
|
||
|
|
"step": 179
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.488135593220339,
|
||
|
|
"grad_norm": 0.390625,
|
||
|
|
"learning_rate": 5.219780219780221e-06,
|
||
|
|
"loss": 0.2376,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.488135593220339,
|
||
|
|
"eval_loss": 0.2105218917131424,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3696,
|
||
|
|
"eval_samples_per_second": 8.795,
|
||
|
|
"eval_steps_per_second": 4.398,
|
||
|
|
"step": 180
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4908474576271186,
|
||
|
|
"grad_norm": 0.375,
|
||
|
|
"learning_rate": 5.192307692307693e-06,
|
||
|
|
"loss": 0.1714,
|
||
|
|
"step": 181
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.4935593220338983,
|
||
|
|
"grad_norm": 0.376953125,
|
||
|
|
"learning_rate": 5.164835164835166e-06,
|
||
|
|
"loss": 0.1925,
|
||
|
|
"step": 182
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.49627118644067797,
|
||
|
|
"grad_norm": 0.37890625,
|
||
|
|
"learning_rate": 5.137362637362638e-06,
|
||
|
|
"loss": 0.1919,
|
||
|
|
"step": 183
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.49898305084745764,
|
||
|
|
"grad_norm": 0.546875,
|
||
|
|
"learning_rate": 5.10989010989011e-06,
|
||
|
|
"loss": 0.2434,
|
||
|
|
"step": 184
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5016949152542373,
|
||
|
|
"grad_norm": 0.314453125,
|
||
|
|
"learning_rate": 5.0824175824175824e-06,
|
||
|
|
"loss": 0.1381,
|
||
|
|
"step": 185
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.504406779661017,
|
||
|
|
"grad_norm": 0.482421875,
|
||
|
|
"learning_rate": 5.054945054945055e-06,
|
||
|
|
"loss": 0.2796,
|
||
|
|
"step": 186
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5071186440677966,
|
||
|
|
"grad_norm": 0.40234375,
|
||
|
|
"learning_rate": 5.027472527472528e-06,
|
||
|
|
"loss": 0.1945,
|
||
|
|
"step": 187
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5098305084745762,
|
||
|
|
"grad_norm": 0.431640625,
|
||
|
|
"learning_rate": 5e-06,
|
||
|
|
"loss": 0.2571,
|
||
|
|
"step": 188
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.512542372881356,
|
||
|
|
"grad_norm": 0.455078125,
|
||
|
|
"learning_rate": 4.972527472527473e-06,
|
||
|
|
"loss": 0.2749,
|
||
|
|
"step": 189
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5152542372881356,
|
||
|
|
"grad_norm": 0.42578125,
|
||
|
|
"learning_rate": 4.945054945054946e-06,
|
||
|
|
"loss": 0.2185,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5152542372881356,
|
||
|
|
"eval_loss": 0.2096569836139679,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3456,
|
||
|
|
"eval_samples_per_second": 8.814,
|
||
|
|
"eval_steps_per_second": 4.407,
|
||
|
|
"step": 190
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5179661016949153,
|
||
|
|
"grad_norm": 0.404296875,
|
||
|
|
"learning_rate": 4.9175824175824175e-06,
|
||
|
|
"loss": 0.2161,
|
||
|
|
"step": 191
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5206779661016949,
|
||
|
|
"grad_norm": 0.421875,
|
||
|
|
"learning_rate": 4.890109890109891e-06,
|
||
|
|
"loss": 0.237,
|
||
|
|
"step": 192
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5233898305084745,
|
||
|
|
"grad_norm": 0.40234375,
|
||
|
|
"learning_rate": 4.862637362637363e-06,
|
||
|
|
"loss": 0.2074,
|
||
|
|
"step": 193
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5261016949152543,
|
||
|
|
"grad_norm": 0.373046875,
|
||
|
|
"learning_rate": 4.8351648351648355e-06,
|
||
|
|
"loss": 0.2588,
|
||
|
|
"step": 194
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5288135593220339,
|
||
|
|
"grad_norm": 0.451171875,
|
||
|
|
"learning_rate": 4.807692307692308e-06,
|
||
|
|
"loss": 0.2358,
|
||
|
|
"step": 195
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5315254237288135,
|
||
|
|
"grad_norm": 0.3046875,
|
||
|
|
"learning_rate": 4.780219780219781e-06,
|
||
|
|
"loss": 0.1249,
|
||
|
|
"step": 196
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5342372881355932,
|
||
|
|
"grad_norm": 0.41015625,
|
||
|
|
"learning_rate": 4.752747252747253e-06,
|
||
|
|
"loss": 0.1959,
|
||
|
|
"step": 197
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5369491525423729,
|
||
|
|
"grad_norm": 0.2451171875,
|
||
|
|
"learning_rate": 4.725274725274726e-06,
|
||
|
|
"loss": 0.1217,
|
||
|
|
"step": 198
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5396610169491526,
|
||
|
|
"grad_norm": 0.310546875,
|
||
|
|
"learning_rate": 4.697802197802198e-06,
|
||
|
|
"loss": 0.1453,
|
||
|
|
"step": 199
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5423728813559322,
|
||
|
|
"grad_norm": 0.416015625,
|
||
|
|
"learning_rate": 4.6703296703296706e-06,
|
||
|
|
"loss": 0.2259,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5423728813559322,
|
||
|
|
"eval_loss": 0.20834746956825256,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3813,
|
||
|
|
"eval_samples_per_second": 8.786,
|
||
|
|
"eval_steps_per_second": 4.393,
|
||
|
|
"step": 200
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5450847457627118,
|
||
|
|
"grad_norm": 0.408203125,
|
||
|
|
"learning_rate": 4.642857142857144e-06,
|
||
|
|
"loss": 0.1652,
|
||
|
|
"step": 201
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5477966101694915,
|
||
|
|
"grad_norm": 0.392578125,
|
||
|
|
"learning_rate": 4.615384615384616e-06,
|
||
|
|
"loss": 0.2135,
|
||
|
|
"step": 202
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5505084745762712,
|
||
|
|
"grad_norm": 0.365234375,
|
||
|
|
"learning_rate": 4.587912087912088e-06,
|
||
|
|
"loss": 0.2018,
|
||
|
|
"step": 203
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5532203389830509,
|
||
|
|
"grad_norm": 0.330078125,
|
||
|
|
"learning_rate": 4.560439560439561e-06,
|
||
|
|
"loss": 0.1461,
|
||
|
|
"step": 204
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5559322033898305,
|
||
|
|
"grad_norm": 0.4609375,
|
||
|
|
"learning_rate": 4.532967032967033e-06,
|
||
|
|
"loss": 0.2551,
|
||
|
|
"step": 205
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5586440677966101,
|
||
|
|
"grad_norm": 0.423828125,
|
||
|
|
"learning_rate": 4.505494505494506e-06,
|
||
|
|
"loss": 0.1913,
|
||
|
|
"step": 206
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5613559322033899,
|
||
|
|
"grad_norm": 0.4375,
|
||
|
|
"learning_rate": 4.478021978021979e-06,
|
||
|
|
"loss": 0.2274,
|
||
|
|
"step": 207
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5640677966101695,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 4.45054945054945e-06,
|
||
|
|
"loss": 0.1962,
|
||
|
|
"step": 208
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5667796610169491,
|
||
|
|
"grad_norm": 0.494140625,
|
||
|
|
"learning_rate": 4.423076923076924e-06,
|
||
|
|
"loss": 0.2946,
|
||
|
|
"step": 209
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5694915254237288,
|
||
|
|
"grad_norm": 0.4609375,
|
||
|
|
"learning_rate": 4.395604395604396e-06,
|
||
|
|
"loss": 0.2301,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5694915254237288,
|
||
|
|
"eval_loss": 0.20648249983787537,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3454,
|
||
|
|
"eval_samples_per_second": 8.814,
|
||
|
|
"eval_steps_per_second": 4.407,
|
||
|
|
"step": 210
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5722033898305084,
|
||
|
|
"grad_norm": 0.39453125,
|
||
|
|
"learning_rate": 4.368131868131868e-06,
|
||
|
|
"loss": 0.2761,
|
||
|
|
"step": 211
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5749152542372882,
|
||
|
|
"grad_norm": 0.306640625,
|
||
|
|
"learning_rate": 4.340659340659341e-06,
|
||
|
|
"loss": 0.1019,
|
||
|
|
"step": 212
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5776271186440678,
|
||
|
|
"grad_norm": 0.51953125,
|
||
|
|
"learning_rate": 4.313186813186814e-06,
|
||
|
|
"loss": 0.3189,
|
||
|
|
"step": 213
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5803389830508474,
|
||
|
|
"grad_norm": 0.40625,
|
||
|
|
"learning_rate": 4.2857142857142855e-06,
|
||
|
|
"loss": 0.222,
|
||
|
|
"step": 214
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5830508474576271,
|
||
|
|
"grad_norm": 0.45703125,
|
||
|
|
"learning_rate": 4.258241758241759e-06,
|
||
|
|
"loss": 0.271,
|
||
|
|
"step": 215
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5857627118644068,
|
||
|
|
"grad_norm": 0.421875,
|
||
|
|
"learning_rate": 4.230769230769231e-06,
|
||
|
|
"loss": 0.2659,
|
||
|
|
"step": 216
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5884745762711865,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 4.203296703296703e-06,
|
||
|
|
"loss": 0.1449,
|
||
|
|
"step": 217
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5911864406779661,
|
||
|
|
"grad_norm": 0.482421875,
|
||
|
|
"learning_rate": 4.175824175824177e-06,
|
||
|
|
"loss": 0.3675,
|
||
|
|
"step": 218
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5938983050847457,
|
||
|
|
"grad_norm": 0.423828125,
|
||
|
|
"learning_rate": 4.148351648351649e-06,
|
||
|
|
"loss": 0.2713,
|
||
|
|
"step": 219
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5966101694915255,
|
||
|
|
"grad_norm": 0.357421875,
|
||
|
|
"learning_rate": 4.120879120879121e-06,
|
||
|
|
"loss": 0.1717,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5966101694915255,
|
||
|
|
"eval_loss": 0.20527859032154083,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3389,
|
||
|
|
"eval_samples_per_second": 8.819,
|
||
|
|
"eval_steps_per_second": 4.41,
|
||
|
|
"step": 220
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.5993220338983051,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 4.093406593406594e-06,
|
||
|
|
"loss": 0.2136,
|
||
|
|
"step": 221
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6020338983050847,
|
||
|
|
"grad_norm": 0.259765625,
|
||
|
|
"learning_rate": 4.065934065934066e-06,
|
||
|
|
"loss": 0.1038,
|
||
|
|
"step": 222
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6047457627118644,
|
||
|
|
"grad_norm": 0.345703125,
|
||
|
|
"learning_rate": 4.0384615384615385e-06,
|
||
|
|
"loss": 0.1795,
|
||
|
|
"step": 223
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.607457627118644,
|
||
|
|
"grad_norm": 0.359375,
|
||
|
|
"learning_rate": 4.010989010989012e-06,
|
||
|
|
"loss": 0.2246,
|
||
|
|
"step": 224
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6101694915254238,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 3.983516483516483e-06,
|
||
|
|
"loss": 0.2252,
|
||
|
|
"step": 225
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6128813559322034,
|
||
|
|
"grad_norm": 0.396484375,
|
||
|
|
"learning_rate": 3.9560439560439565e-06,
|
||
|
|
"loss": 0.2359,
|
||
|
|
"step": 226
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.615593220338983,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 3.928571428571429e-06,
|
||
|
|
"loss": 0.1369,
|
||
|
|
"step": 227
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6183050847457627,
|
||
|
|
"grad_norm": 0.43359375,
|
||
|
|
"learning_rate": 3.901098901098901e-06,
|
||
|
|
"loss": 0.2164,
|
||
|
|
"step": 228
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6210169491525424,
|
||
|
|
"grad_norm": 0.37890625,
|
||
|
|
"learning_rate": 3.873626373626374e-06,
|
||
|
|
"loss": 0.1927,
|
||
|
|
"step": 229
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6237288135593221,
|
||
|
|
"grad_norm": 0.341796875,
|
||
|
|
"learning_rate": 3.846153846153847e-06,
|
||
|
|
"loss": 0.1609,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6237288135593221,
|
||
|
|
"eval_loss": 0.20454387366771698,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.4641,
|
||
|
|
"eval_samples_per_second": 8.723,
|
||
|
|
"eval_steps_per_second": 4.361,
|
||
|
|
"step": 230
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6264406779661017,
|
||
|
|
"grad_norm": 0.41796875,
|
||
|
|
"learning_rate": 3.818681318681319e-06,
|
||
|
|
"loss": 0.2662,
|
||
|
|
"step": 231
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6291525423728813,
|
||
|
|
"grad_norm": 0.306640625,
|
||
|
|
"learning_rate": 3.7912087912087915e-06,
|
||
|
|
"loss": 0.1026,
|
||
|
|
"step": 232
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.631864406779661,
|
||
|
|
"grad_norm": 0.30859375,
|
||
|
|
"learning_rate": 3.763736263736264e-06,
|
||
|
|
"loss": 0.1002,
|
||
|
|
"step": 233
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6345762711864407,
|
||
|
|
"grad_norm": 0.408203125,
|
||
|
|
"learning_rate": 3.7362637362637367e-06,
|
||
|
|
"loss": 0.1949,
|
||
|
|
"step": 234
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6372881355932203,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 3.708791208791209e-06,
|
||
|
|
"loss": 0.2105,
|
||
|
|
"step": 235
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.64,
|
||
|
|
"grad_norm": 0.431640625,
|
||
|
|
"learning_rate": 3.681318681318682e-06,
|
||
|
|
"loss": 0.24,
|
||
|
|
"step": 236
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6427118644067796,
|
||
|
|
"grad_norm": 0.48046875,
|
||
|
|
"learning_rate": 3.653846153846154e-06,
|
||
|
|
"loss": 0.2764,
|
||
|
|
"step": 237
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6454237288135594,
|
||
|
|
"grad_norm": 0.33203125,
|
||
|
|
"learning_rate": 3.6263736263736266e-06,
|
||
|
|
"loss": 0.1852,
|
||
|
|
"step": 238
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.648135593220339,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 3.598901098901099e-06,
|
||
|
|
"loss": 0.1494,
|
||
|
|
"step": 239
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6508474576271186,
|
||
|
|
"grad_norm": 0.328125,
|
||
|
|
"learning_rate": 3.5714285714285718e-06,
|
||
|
|
"loss": 0.1097,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6508474576271186,
|
||
|
|
"eval_loss": 0.2033514529466629,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5749,
|
||
|
|
"eval_samples_per_second": 8.639,
|
||
|
|
"eval_steps_per_second": 4.32,
|
||
|
|
"step": 240
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6535593220338983,
|
||
|
|
"grad_norm": 0.349609375,
|
||
|
|
"learning_rate": 3.5439560439560446e-06,
|
||
|
|
"loss": 0.1553,
|
||
|
|
"step": 241
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.656271186440678,
|
||
|
|
"grad_norm": 0.310546875,
|
||
|
|
"learning_rate": 3.516483516483517e-06,
|
||
|
|
"loss": 0.181,
|
||
|
|
"step": 242
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6589830508474577,
|
||
|
|
"grad_norm": 0.384765625,
|
||
|
|
"learning_rate": 3.4890109890109893e-06,
|
||
|
|
"loss": 0.171,
|
||
|
|
"step": 243
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6616949152542373,
|
||
|
|
"grad_norm": 0.357421875,
|
||
|
|
"learning_rate": 3.4615384615384617e-06,
|
||
|
|
"loss": 0.1853,
|
||
|
|
"step": 244
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6644067796610169,
|
||
|
|
"grad_norm": 0.53125,
|
||
|
|
"learning_rate": 3.4340659340659345e-06,
|
||
|
|
"loss": 0.2538,
|
||
|
|
"step": 245
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6671186440677966,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 3.406593406593407e-06,
|
||
|
|
"loss": 0.1439,
|
||
|
|
"step": 246
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6698305084745763,
|
||
|
|
"grad_norm": 0.3515625,
|
||
|
|
"learning_rate": 3.3791208791208797e-06,
|
||
|
|
"loss": 0.1826,
|
||
|
|
"step": 247
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.672542372881356,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 3.3516483516483516e-06,
|
||
|
|
"loss": 0.2208,
|
||
|
|
"step": 248
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6752542372881356,
|
||
|
|
"grad_norm": 0.365234375,
|
||
|
|
"learning_rate": 3.3241758241758244e-06,
|
||
|
|
"loss": 0.1267,
|
||
|
|
"step": 249
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6779661016949152,
|
||
|
|
"grad_norm": 0.345703125,
|
||
|
|
"learning_rate": 3.2967032967032968e-06,
|
||
|
|
"loss": 0.1361,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6779661016949152,
|
||
|
|
"eval_loss": 0.20349906384944916,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5599,
|
||
|
|
"eval_samples_per_second": 8.651,
|
||
|
|
"eval_steps_per_second": 4.325,
|
||
|
|
"step": 250
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.680677966101695,
|
||
|
|
"grad_norm": 0.478515625,
|
||
|
|
"learning_rate": 3.2692307692307696e-06,
|
||
|
|
"loss": 0.3215,
|
||
|
|
"step": 251
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6833898305084746,
|
||
|
|
"grad_norm": 0.455078125,
|
||
|
|
"learning_rate": 3.2417582417582424e-06,
|
||
|
|
"loss": 0.2283,
|
||
|
|
"step": 252
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6861016949152542,
|
||
|
|
"grad_norm": 0.404296875,
|
||
|
|
"learning_rate": 3.2142857142857147e-06,
|
||
|
|
"loss": 0.179,
|
||
|
|
"step": 253
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6888135593220339,
|
||
|
|
"grad_norm": 0.33203125,
|
||
|
|
"learning_rate": 3.1868131868131867e-06,
|
||
|
|
"loss": 0.1262,
|
||
|
|
"step": 254
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6915254237288135,
|
||
|
|
"grad_norm": 0.34375,
|
||
|
|
"learning_rate": 3.1593406593406595e-06,
|
||
|
|
"loss": 0.1797,
|
||
|
|
"step": 255
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6942372881355933,
|
||
|
|
"grad_norm": 0.453125,
|
||
|
|
"learning_rate": 3.1318681318681323e-06,
|
||
|
|
"loss": 0.2528,
|
||
|
|
"step": 256
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6969491525423729,
|
||
|
|
"grad_norm": 0.404296875,
|
||
|
|
"learning_rate": 3.1043956043956046e-06,
|
||
|
|
"loss": 0.1432,
|
||
|
|
"step": 257
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.6996610169491525,
|
||
|
|
"grad_norm": 0.35546875,
|
||
|
|
"learning_rate": 3.0769230769230774e-06,
|
||
|
|
"loss": 0.1417,
|
||
|
|
"step": 258
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7023728813559322,
|
||
|
|
"grad_norm": 0.443359375,
|
||
|
|
"learning_rate": 3.04945054945055e-06,
|
||
|
|
"loss": 0.2974,
|
||
|
|
"step": 259
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7050847457627119,
|
||
|
|
"grad_norm": 0.361328125,
|
||
|
|
"learning_rate": 3.021978021978022e-06,
|
||
|
|
"loss": 0.2162,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7050847457627119,
|
||
|
|
"eval_loss": 0.20323127508163452,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5157,
|
||
|
|
"eval_samples_per_second": 8.684,
|
||
|
|
"eval_steps_per_second": 4.342,
|
||
|
|
"step": 260
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7077966101694916,
|
||
|
|
"grad_norm": 0.412109375,
|
||
|
|
"learning_rate": 2.9945054945054946e-06,
|
||
|
|
"loss": 0.1693,
|
||
|
|
"step": 261
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7105084745762712,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 2.9670329670329673e-06,
|
||
|
|
"loss": 0.1286,
|
||
|
|
"step": 262
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7132203389830508,
|
||
|
|
"grad_norm": 0.380859375,
|
||
|
|
"learning_rate": 2.9395604395604397e-06,
|
||
|
|
"loss": 0.1939,
|
||
|
|
"step": 263
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7159322033898305,
|
||
|
|
"grad_norm": 0.287109375,
|
||
|
|
"learning_rate": 2.9120879120879125e-06,
|
||
|
|
"loss": 0.1059,
|
||
|
|
"step": 264
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7186440677966102,
|
||
|
|
"grad_norm": 0.4296875,
|
||
|
|
"learning_rate": 2.8846153846153845e-06,
|
||
|
|
"loss": 0.1912,
|
||
|
|
"step": 265
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7213559322033898,
|
||
|
|
"grad_norm": 0.318359375,
|
||
|
|
"learning_rate": 2.8571428571428573e-06,
|
||
|
|
"loss": 0.1356,
|
||
|
|
"step": 266
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7240677966101695,
|
||
|
|
"grad_norm": 0.333984375,
|
||
|
|
"learning_rate": 2.82967032967033e-06,
|
||
|
|
"loss": 0.1228,
|
||
|
|
"step": 267
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7267796610169491,
|
||
|
|
"grad_norm": 0.37109375,
|
||
|
|
"learning_rate": 2.8021978021978024e-06,
|
||
|
|
"loss": 0.2007,
|
||
|
|
"step": 268
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7294915254237289,
|
||
|
|
"grad_norm": 0.37890625,
|
||
|
|
"learning_rate": 2.7747252747252752e-06,
|
||
|
|
"loss": 0.2002,
|
||
|
|
"step": 269
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7322033898305085,
|
||
|
|
"grad_norm": 0.3984375,
|
||
|
|
"learning_rate": 2.7472527472527476e-06,
|
||
|
|
"loss": 0.1859,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7322033898305085,
|
||
|
|
"eval_loss": 0.20238856971263885,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5157,
|
||
|
|
"eval_samples_per_second": 8.684,
|
||
|
|
"eval_steps_per_second": 4.342,
|
||
|
|
"step": 270
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7349152542372881,
|
||
|
|
"grad_norm": 0.359375,
|
||
|
|
"learning_rate": 2.71978021978022e-06,
|
||
|
|
"loss": 0.1753,
|
||
|
|
"step": 271
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7376271186440678,
|
||
|
|
"grad_norm": 0.484375,
|
||
|
|
"learning_rate": 2.6923076923076923e-06,
|
||
|
|
"loss": 0.1885,
|
||
|
|
"step": 272
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7403389830508474,
|
||
|
|
"grad_norm": 0.416015625,
|
||
|
|
"learning_rate": 2.664835164835165e-06,
|
||
|
|
"loss": 0.2015,
|
||
|
|
"step": 273
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7430508474576272,
|
||
|
|
"grad_norm": 0.3359375,
|
||
|
|
"learning_rate": 2.6373626373626375e-06,
|
||
|
|
"loss": 0.1205,
|
||
|
|
"step": 274
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7457627118644068,
|
||
|
|
"grad_norm": 0.404296875,
|
||
|
|
"learning_rate": 2.6098901098901103e-06,
|
||
|
|
"loss": 0.1731,
|
||
|
|
"step": 275
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7484745762711864,
|
||
|
|
"grad_norm": 0.4609375,
|
||
|
|
"learning_rate": 2.582417582417583e-06,
|
||
|
|
"loss": 0.2409,
|
||
|
|
"step": 276
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7511864406779661,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 2.554945054945055e-06,
|
||
|
|
"loss": 0.2074,
|
||
|
|
"step": 277
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7538983050847458,
|
||
|
|
"grad_norm": 0.51953125,
|
||
|
|
"learning_rate": 2.5274725274725274e-06,
|
||
|
|
"loss": 0.3189,
|
||
|
|
"step": 278
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7566101694915254,
|
||
|
|
"grad_norm": 0.322265625,
|
||
|
|
"learning_rate": 2.5e-06,
|
||
|
|
"loss": 0.1501,
|
||
|
|
"step": 279
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7593220338983051,
|
||
|
|
"grad_norm": 0.396484375,
|
||
|
|
"learning_rate": 2.472527472527473e-06,
|
||
|
|
"loss": 0.209,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7593220338983051,
|
||
|
|
"eval_loss": 0.2017662674188614,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5035,
|
||
|
|
"eval_samples_per_second": 8.693,
|
||
|
|
"eval_steps_per_second": 4.347,
|
||
|
|
"step": 280
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7620338983050847,
|
||
|
|
"grad_norm": 0.375,
|
||
|
|
"learning_rate": 2.4450549450549454e-06,
|
||
|
|
"loss": 0.178,
|
||
|
|
"step": 281
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7647457627118645,
|
||
|
|
"grad_norm": 0.408203125,
|
||
|
|
"learning_rate": 2.4175824175824177e-06,
|
||
|
|
"loss": 0.2037,
|
||
|
|
"step": 282
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7674576271186441,
|
||
|
|
"grad_norm": 0.455078125,
|
||
|
|
"learning_rate": 2.3901098901098905e-06,
|
||
|
|
"loss": 0.2565,
|
||
|
|
"step": 283
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7701694915254237,
|
||
|
|
"grad_norm": 0.451171875,
|
||
|
|
"learning_rate": 2.362637362637363e-06,
|
||
|
|
"loss": 0.2449,
|
||
|
|
"step": 284
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7728813559322034,
|
||
|
|
"grad_norm": 0.486328125,
|
||
|
|
"learning_rate": 2.3351648351648353e-06,
|
||
|
|
"loss": 0.2613,
|
||
|
|
"step": 285
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.775593220338983,
|
||
|
|
"grad_norm": 0.38671875,
|
||
|
|
"learning_rate": 2.307692307692308e-06,
|
||
|
|
"loss": 0.2347,
|
||
|
|
"step": 286
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7783050847457628,
|
||
|
|
"grad_norm": 0.4140625,
|
||
|
|
"learning_rate": 2.2802197802197804e-06,
|
||
|
|
"loss": 0.1807,
|
||
|
|
"step": 287
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7810169491525424,
|
||
|
|
"grad_norm": 0.388671875,
|
||
|
|
"learning_rate": 2.252747252747253e-06,
|
||
|
|
"loss": 0.2093,
|
||
|
|
"step": 288
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.783728813559322,
|
||
|
|
"grad_norm": 0.38671875,
|
||
|
|
"learning_rate": 2.225274725274725e-06,
|
||
|
|
"loss": 0.1885,
|
||
|
|
"step": 289
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7864406779661017,
|
||
|
|
"grad_norm": 0.296875,
|
||
|
|
"learning_rate": 2.197802197802198e-06,
|
||
|
|
"loss": 0.1163,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7864406779661017,
|
||
|
|
"eval_loss": 0.2012355625629425,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5261,
|
||
|
|
"eval_samples_per_second": 8.676,
|
||
|
|
"eval_steps_per_second": 4.338,
|
||
|
|
"step": 290
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7891525423728813,
|
||
|
|
"grad_norm": 0.484375,
|
||
|
|
"learning_rate": 2.1703296703296704e-06,
|
||
|
|
"loss": 0.2859,
|
||
|
|
"step": 291
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.791864406779661,
|
||
|
|
"grad_norm": 0.37890625,
|
||
|
|
"learning_rate": 2.1428571428571427e-06,
|
||
|
|
"loss": 0.186,
|
||
|
|
"step": 292
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7945762711864407,
|
||
|
|
"grad_norm": 0.36328125,
|
||
|
|
"learning_rate": 2.1153846153846155e-06,
|
||
|
|
"loss": 0.1688,
|
||
|
|
"step": 293
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.7972881355932203,
|
||
|
|
"grad_norm": 0.291015625,
|
||
|
|
"learning_rate": 2.0879120879120883e-06,
|
||
|
|
"loss": 0.1007,
|
||
|
|
"step": 294
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8,
|
||
|
|
"grad_norm": 0.53515625,
|
||
|
|
"learning_rate": 2.0604395604395607e-06,
|
||
|
|
"loss": 0.3142,
|
||
|
|
"step": 295
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8027118644067797,
|
||
|
|
"grad_norm": 0.439453125,
|
||
|
|
"learning_rate": 2.032967032967033e-06,
|
||
|
|
"loss": 0.2239,
|
||
|
|
"step": 296
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8054237288135593,
|
||
|
|
"grad_norm": 0.400390625,
|
||
|
|
"learning_rate": 2.005494505494506e-06,
|
||
|
|
"loss": 0.1749,
|
||
|
|
"step": 297
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.808135593220339,
|
||
|
|
"grad_norm": 0.439453125,
|
||
|
|
"learning_rate": 1.9780219780219782e-06,
|
||
|
|
"loss": 0.2692,
|
||
|
|
"step": 298
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8108474576271186,
|
||
|
|
"grad_norm": 0.466796875,
|
||
|
|
"learning_rate": 1.9505494505494506e-06,
|
||
|
|
"loss": 0.3106,
|
||
|
|
"step": 299
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8135593220338984,
|
||
|
|
"grad_norm": 0.359375,
|
||
|
|
"learning_rate": 1.9230769230769234e-06,
|
||
|
|
"loss": 0.1723,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8135593220338984,
|
||
|
|
"eval_loss": 0.20140354335308075,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5168,
|
||
|
|
"eval_samples_per_second": 8.683,
|
||
|
|
"eval_steps_per_second": 4.342,
|
||
|
|
"step": 300
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.816271186440678,
|
||
|
|
"grad_norm": 0.45703125,
|
||
|
|
"learning_rate": 1.8956043956043958e-06,
|
||
|
|
"loss": 0.1482,
|
||
|
|
"step": 301
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8189830508474576,
|
||
|
|
"grad_norm": 0.4375,
|
||
|
|
"learning_rate": 1.8681318681318684e-06,
|
||
|
|
"loss": 0.2671,
|
||
|
|
"step": 302
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8216949152542373,
|
||
|
|
"grad_norm": 0.421875,
|
||
|
|
"learning_rate": 1.840659340659341e-06,
|
||
|
|
"loss": 0.2112,
|
||
|
|
"step": 303
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8244067796610169,
|
||
|
|
"grad_norm": 0.4140625,
|
||
|
|
"learning_rate": 1.8131868131868133e-06,
|
||
|
|
"loss": 0.1977,
|
||
|
|
"step": 304
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8271186440677966,
|
||
|
|
"grad_norm": 0.3046875,
|
||
|
|
"learning_rate": 1.7857142857142859e-06,
|
||
|
|
"loss": 0.1583,
|
||
|
|
"step": 305
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8298305084745763,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 1.7582417582417585e-06,
|
||
|
|
"loss": 0.1156,
|
||
|
|
"step": 306
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8325423728813559,
|
||
|
|
"grad_norm": 0.322265625,
|
||
|
|
"learning_rate": 1.7307692307692308e-06,
|
||
|
|
"loss": 0.1492,
|
||
|
|
"step": 307
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8352542372881356,
|
||
|
|
"grad_norm": 0.4609375,
|
||
|
|
"learning_rate": 1.7032967032967034e-06,
|
||
|
|
"loss": 0.2713,
|
||
|
|
"step": 308
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8379661016949153,
|
||
|
|
"grad_norm": 0.384765625,
|
||
|
|
"learning_rate": 1.6758241758241758e-06,
|
||
|
|
"loss": 0.1979,
|
||
|
|
"step": 309
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8406779661016949,
|
||
|
|
"grad_norm": 0.30859375,
|
||
|
|
"learning_rate": 1.6483516483516484e-06,
|
||
|
|
"loss": 0.1339,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8406779661016949,
|
||
|
|
"eval_loss": 0.2008802443742752,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5181,
|
||
|
|
"eval_samples_per_second": 8.682,
|
||
|
|
"eval_steps_per_second": 4.341,
|
||
|
|
"step": 310
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8433898305084746,
|
||
|
|
"grad_norm": 0.435546875,
|
||
|
|
"learning_rate": 1.6208791208791212e-06,
|
||
|
|
"loss": 0.1836,
|
||
|
|
"step": 311
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8461016949152542,
|
||
|
|
"grad_norm": 0.326171875,
|
||
|
|
"learning_rate": 1.5934065934065933e-06,
|
||
|
|
"loss": 0.1229,
|
||
|
|
"step": 312
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.848813559322034,
|
||
|
|
"grad_norm": 0.41015625,
|
||
|
|
"learning_rate": 1.5659340659340661e-06,
|
||
|
|
"loss": 0.2302,
|
||
|
|
"step": 313
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8515254237288136,
|
||
|
|
"grad_norm": 0.470703125,
|
||
|
|
"learning_rate": 1.5384615384615387e-06,
|
||
|
|
"loss": 0.283,
|
||
|
|
"step": 314
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8542372881355932,
|
||
|
|
"grad_norm": 0.41796875,
|
||
|
|
"learning_rate": 1.510989010989011e-06,
|
||
|
|
"loss": 0.1891,
|
||
|
|
"step": 315
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8569491525423729,
|
||
|
|
"grad_norm": 0.361328125,
|
||
|
|
"learning_rate": 1.4835164835164837e-06,
|
||
|
|
"loss": 0.2037,
|
||
|
|
"step": 316
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8596610169491525,
|
||
|
|
"grad_norm": 0.3984375,
|
||
|
|
"learning_rate": 1.4560439560439563e-06,
|
||
|
|
"loss": 0.1894,
|
||
|
|
"step": 317
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8623728813559322,
|
||
|
|
"grad_norm": 0.37890625,
|
||
|
|
"learning_rate": 1.4285714285714286e-06,
|
||
|
|
"loss": 0.188,
|
||
|
|
"step": 318
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8650847457627119,
|
||
|
|
"grad_norm": 0.3984375,
|
||
|
|
"learning_rate": 1.4010989010989012e-06,
|
||
|
|
"loss": 0.1888,
|
||
|
|
"step": 319
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8677966101694915,
|
||
|
|
"grad_norm": 0.3515625,
|
||
|
|
"learning_rate": 1.3736263736263738e-06,
|
||
|
|
"loss": 0.136,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8677966101694915,
|
||
|
|
"eval_loss": 0.20059189200401306,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5219,
|
||
|
|
"eval_samples_per_second": 8.679,
|
||
|
|
"eval_steps_per_second": 4.34,
|
||
|
|
"step": 320
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8705084745762712,
|
||
|
|
"grad_norm": 0.44140625,
|
||
|
|
"learning_rate": 1.3461538461538462e-06,
|
||
|
|
"loss": 0.1919,
|
||
|
|
"step": 321
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8732203389830508,
|
||
|
|
"grad_norm": 0.4140625,
|
||
|
|
"learning_rate": 1.3186813186813187e-06,
|
||
|
|
"loss": 0.1494,
|
||
|
|
"step": 322
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8759322033898305,
|
||
|
|
"grad_norm": 0.40625,
|
||
|
|
"learning_rate": 1.2912087912087915e-06,
|
||
|
|
"loss": 0.1777,
|
||
|
|
"step": 323
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8786440677966102,
|
||
|
|
"grad_norm": 0.29296875,
|
||
|
|
"learning_rate": 1.2637362637362637e-06,
|
||
|
|
"loss": 0.1137,
|
||
|
|
"step": 324
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8813559322033898,
|
||
|
|
"grad_norm": 0.53125,
|
||
|
|
"learning_rate": 1.2362637362637365e-06,
|
||
|
|
"loss": 0.2659,
|
||
|
|
"step": 325
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8840677966101695,
|
||
|
|
"grad_norm": 0.46875,
|
||
|
|
"learning_rate": 1.2087912087912089e-06,
|
||
|
|
"loss": 0.2715,
|
||
|
|
"step": 326
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8867796610169492,
|
||
|
|
"grad_norm": 0.53515625,
|
||
|
|
"learning_rate": 1.1813186813186815e-06,
|
||
|
|
"loss": 0.3199,
|
||
|
|
"step": 327
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8894915254237288,
|
||
|
|
"grad_norm": 0.328125,
|
||
|
|
"learning_rate": 1.153846153846154e-06,
|
||
|
|
"loss": 0.1237,
|
||
|
|
"step": 328
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8922033898305085,
|
||
|
|
"grad_norm": 0.392578125,
|
||
|
|
"learning_rate": 1.1263736263736264e-06,
|
||
|
|
"loss": 0.1683,
|
||
|
|
"step": 329
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8949152542372881,
|
||
|
|
"grad_norm": 0.328125,
|
||
|
|
"learning_rate": 1.098901098901099e-06,
|
||
|
|
"loss": 0.1182,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8949152542372881,
|
||
|
|
"eval_loss": 0.2006920427083969,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5312,
|
||
|
|
"eval_samples_per_second": 8.672,
|
||
|
|
"eval_steps_per_second": 4.336,
|
||
|
|
"step": 330
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.8976271186440677,
|
||
|
|
"grad_norm": 0.384765625,
|
||
|
|
"learning_rate": 1.0714285714285714e-06,
|
||
|
|
"loss": 0.2004,
|
||
|
|
"step": 331
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9003389830508475,
|
||
|
|
"grad_norm": 0.4375,
|
||
|
|
"learning_rate": 1.0439560439560442e-06,
|
||
|
|
"loss": 0.1806,
|
||
|
|
"step": 332
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9030508474576271,
|
||
|
|
"grad_norm": 0.33203125,
|
||
|
|
"learning_rate": 1.0164835164835165e-06,
|
||
|
|
"loss": 0.1207,
|
||
|
|
"step": 333
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9057627118644068,
|
||
|
|
"grad_norm": 0.52734375,
|
||
|
|
"learning_rate": 9.890109890109891e-07,
|
||
|
|
"loss": 0.262,
|
||
|
|
"step": 334
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9084745762711864,
|
||
|
|
"grad_norm": 0.427734375,
|
||
|
|
"learning_rate": 9.615384615384617e-07,
|
||
|
|
"loss": 0.1493,
|
||
|
|
"step": 335
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9111864406779661,
|
||
|
|
"grad_norm": 0.46484375,
|
||
|
|
"learning_rate": 9.340659340659342e-07,
|
||
|
|
"loss": 0.2196,
|
||
|
|
"step": 336
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9138983050847458,
|
||
|
|
"grad_norm": 0.43359375,
|
||
|
|
"learning_rate": 9.065934065934067e-07,
|
||
|
|
"loss": 0.2253,
|
||
|
|
"step": 337
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9166101694915254,
|
||
|
|
"grad_norm": 0.55859375,
|
||
|
|
"learning_rate": 8.791208791208792e-07,
|
||
|
|
"loss": 0.3115,
|
||
|
|
"step": 338
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9193220338983051,
|
||
|
|
"grad_norm": 0.400390625,
|
||
|
|
"learning_rate": 8.516483516483517e-07,
|
||
|
|
"loss": 0.1799,
|
||
|
|
"step": 339
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9220338983050848,
|
||
|
|
"grad_norm": 0.443359375,
|
||
|
|
"learning_rate": 8.241758241758242e-07,
|
||
|
|
"loss": 0.2435,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9220338983050848,
|
||
|
|
"eval_loss": 0.20056262612342834,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.5199,
|
||
|
|
"eval_samples_per_second": 8.681,
|
||
|
|
"eval_steps_per_second": 4.34,
|
||
|
|
"step": 340
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9247457627118644,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 7.967032967032967e-07,
|
||
|
|
"loss": 0.1995,
|
||
|
|
"step": 341
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9274576271186441,
|
||
|
|
"grad_norm": 0.4375,
|
||
|
|
"learning_rate": 7.692307692307694e-07,
|
||
|
|
"loss": 0.2467,
|
||
|
|
"step": 342
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9301694915254237,
|
||
|
|
"grad_norm": 0.447265625,
|
||
|
|
"learning_rate": 7.417582417582418e-07,
|
||
|
|
"loss": 0.1974,
|
||
|
|
"step": 343
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9328813559322033,
|
||
|
|
"grad_norm": 0.455078125,
|
||
|
|
"learning_rate": 7.142857142857143e-07,
|
||
|
|
"loss": 0.1845,
|
||
|
|
"step": 344
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9355932203389831,
|
||
|
|
"grad_norm": 0.4921875,
|
||
|
|
"learning_rate": 6.868131868131869e-07,
|
||
|
|
"loss": 0.2456,
|
||
|
|
"step": 345
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9383050847457627,
|
||
|
|
"grad_norm": 0.37890625,
|
||
|
|
"learning_rate": 6.593406593406594e-07,
|
||
|
|
"loss": 0.2504,
|
||
|
|
"step": 346
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9410169491525424,
|
||
|
|
"grad_norm": 0.416015625,
|
||
|
|
"learning_rate": 6.318681318681319e-07,
|
||
|
|
"loss": 0.2282,
|
||
|
|
"step": 347
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.943728813559322,
|
||
|
|
"grad_norm": 0.474609375,
|
||
|
|
"learning_rate": 6.043956043956044e-07,
|
||
|
|
"loss": 0.2812,
|
||
|
|
"step": 348
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9464406779661017,
|
||
|
|
"grad_norm": 0.29296875,
|
||
|
|
"learning_rate": 5.76923076923077e-07,
|
||
|
|
"loss": 0.111,
|
||
|
|
"step": 349
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9491525423728814,
|
||
|
|
"grad_norm": 0.396484375,
|
||
|
|
"learning_rate": 5.494505494505495e-07,
|
||
|
|
"loss": 0.1889,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9491525423728814,
|
||
|
|
"eval_loss": 0.20049941539764404,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3342,
|
||
|
|
"eval_samples_per_second": 8.823,
|
||
|
|
"eval_steps_per_second": 4.411,
|
||
|
|
"step": 350
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.951864406779661,
|
||
|
|
"grad_norm": 0.47265625,
|
||
|
|
"learning_rate": 5.219780219780221e-07,
|
||
|
|
"loss": 0.2932,
|
||
|
|
"step": 351
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9545762711864407,
|
||
|
|
"grad_norm": 0.291015625,
|
||
|
|
"learning_rate": 4.945054945054946e-07,
|
||
|
|
"loss": 0.139,
|
||
|
|
"step": 352
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9572881355932203,
|
||
|
|
"grad_norm": 0.3828125,
|
||
|
|
"learning_rate": 4.670329670329671e-07,
|
||
|
|
"loss": 0.2242,
|
||
|
|
"step": 353
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.96,
|
||
|
|
"grad_norm": 0.267578125,
|
||
|
|
"learning_rate": 4.395604395604396e-07,
|
||
|
|
"loss": 0.1224,
|
||
|
|
"step": 354
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9627118644067797,
|
||
|
|
"grad_norm": 0.39453125,
|
||
|
|
"learning_rate": 4.120879120879121e-07,
|
||
|
|
"loss": 0.2066,
|
||
|
|
"step": 355
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9654237288135593,
|
||
|
|
"grad_norm": 0.302734375,
|
||
|
|
"learning_rate": 3.846153846153847e-07,
|
||
|
|
"loss": 0.1435,
|
||
|
|
"step": 356
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9681355932203389,
|
||
|
|
"grad_norm": 0.431640625,
|
||
|
|
"learning_rate": 3.5714285714285716e-07,
|
||
|
|
"loss": 0.2338,
|
||
|
|
"step": 357
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9708474576271187,
|
||
|
|
"grad_norm": 0.1826171875,
|
||
|
|
"learning_rate": 3.296703296703297e-07,
|
||
|
|
"loss": 0.0504,
|
||
|
|
"step": 358
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9735593220338983,
|
||
|
|
"grad_norm": 0.419921875,
|
||
|
|
"learning_rate": 3.021978021978022e-07,
|
||
|
|
"loss": 0.2525,
|
||
|
|
"step": 359
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.976271186440678,
|
||
|
|
"grad_norm": 0.41796875,
|
||
|
|
"learning_rate": 2.7472527472527475e-07,
|
||
|
|
"loss": 0.2248,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.976271186440678,
|
||
|
|
"eval_loss": 0.20055252313613892,
|
||
|
|
"eval_model_preparation_time": 0.0393,
|
||
|
|
"eval_runtime": 11.3303,
|
||
|
|
"eval_samples_per_second": 8.826,
|
||
|
|
"eval_steps_per_second": 4.413,
|
||
|
|
"step": 360
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9789830508474576,
|
||
|
|
"grad_norm": 0.353515625,
|
||
|
|
"learning_rate": 2.472527472527473e-07,
|
||
|
|
"loss": 0.1681,
|
||
|
|
"step": 361
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9816949152542372,
|
||
|
|
"grad_norm": 0.35546875,
|
||
|
|
"learning_rate": 2.197802197802198e-07,
|
||
|
|
"loss": 0.1444,
|
||
|
|
"step": 362
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.984406779661017,
|
||
|
|
"grad_norm": 0.392578125,
|
||
|
|
"learning_rate": 1.9230769230769234e-07,
|
||
|
|
"loss": 0.2296,
|
||
|
|
"step": 363
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9871186440677966,
|
||
|
|
"grad_norm": 0.35546875,
|
||
|
|
"learning_rate": 1.6483516483516484e-07,
|
||
|
|
"loss": 0.151,
|
||
|
|
"step": 364
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9898305084745763,
|
||
|
|
"grad_norm": 0.369140625,
|
||
|
|
"learning_rate": 1.3736263736263737e-07,
|
||
|
|
"loss": 0.1958,
|
||
|
|
"step": 365
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9925423728813559,
|
||
|
|
"grad_norm": 0.4453125,
|
||
|
|
"learning_rate": 1.098901098901099e-07,
|
||
|
|
"loss": 0.2799,
|
||
|
|
"step": 366
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9952542372881356,
|
||
|
|
"grad_norm": 0.45703125,
|
||
|
|
"learning_rate": 8.241758241758242e-08,
|
||
|
|
"loss": 0.216,
|
||
|
|
"step": 367
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 0.9979661016949153,
|
||
|
|
"grad_norm": 0.3125,
|
||
|
|
"learning_rate": 5.494505494505495e-08,
|
||
|
|
"loss": 0.146,
|
||
|
|
"step": 368
|
||
|
|
},
|
||
|
|
{
|
||
|
|
"epoch": 1.0,
|
||
|
|
"grad_norm": 0.4921875,
|
||
|
|
"learning_rate": 2.7472527472527476e-08,
|
||
|
|
"loss": 0.1912,
|
||
|
|
"step": 369
|
||
|
|
}
|
||
|
|
],
|
||
|
|
"logging_steps": 1,
|
||
|
|
"max_steps": 369,
|
||
|
|
"num_input_tokens_seen": 0,
|
||
|
|
"num_train_epochs": 1,
|
||
|
|
"save_steps": 5000,
|
||
|
|
"stateful_callbacks": {
|
||
|
|
"TrainerControl": {
|
||
|
|
"args": {
|
||
|
|
"should_epoch_stop": false,
|
||
|
|
"should_evaluate": false,
|
||
|
|
"should_log": false,
|
||
|
|
"should_save": true,
|
||
|
|
"should_training_stop": true
|
||
|
|
},
|
||
|
|
"attributes": {}
|
||
|
|
}
|
||
|
|
},
|
||
|
|
"total_flos": 3.2727396811268506e+17,
|
||
|
|
"train_batch_size": 2,
|
||
|
|
"trial_name": null,
|
||
|
|
"trial_params": null
|
||
|
|
}
|