Cisco1963/llmplasticity-en_zh_linear_0.125_1-seed42
1
0
Fork 0
Code Issues Pull Requests Actions Projects Releases Wiki Activity

初始化项目，由ModelHub XC社区提供模型

Browse Source 
Model: Cisco1963/llmplasticity-en_zh_linear_0.125_1-seed42
Source: Original Platform
This commit is contained in:
ModelHub XC
2026-05-01 06:20:27 +08:00
commit ee5e46d8ff
158 changed files with 4508267 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

194
checkpoint-512/trainer_state.json Normal file
View File

@@ -0,0 +1,194 @@
{
"best_global_step": null,
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.09594415750207887,
"eval_steps": 99999999,
"global_step": 512,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0001873909326212478,
"grad_norm": 245.6398468017578,
"learning_rate": 5e-05,
"loss": 175.7636,
"step": 1
},
{
"epoch": 0.0001873909326212478,
"eval_accuracy": 0.015452299171252725,
"eval_loss": 10.732580184936523,
"eval_runtime": 219.7551,
"eval_samples_per_second": 35.922,
"eval_steps_per_second": 4.491,
"step": 1
},
{
"epoch": 0.0003747818652424956,
"grad_norm": 113.21331024169922,
"learning_rate": 4.9990631440884397e-05,
"loss": 161.0339,
"step": 2
},
{
"epoch": 0.0003747818652424956,
"eval_accuracy": 0.015452299171252725,
"eval_loss": 10.66187572479248,
"eval_runtime": 137.4523,
"eval_samples_per_second": 57.431,
"eval_steps_per_second": 7.181,
"step": 2
},
{
"epoch": 0.0007495637304849911,
"grad_norm": 65.50191497802734,
"learning_rate": 4.997189432265318e-05,
"loss": 156.102,
"step": 4
},
{
"epoch": 0.0007495637304849911,
"eval_accuracy": 0.015464063058397669,
"eval_loss": 10.59908676147461,
"eval_runtime": 188.7788,
"eval_samples_per_second": 41.816,
"eval_steps_per_second": 5.228,
"step": 4
},
{
"epoch": 0.0014991274609699823,
"grad_norm": 38.306678771972656,
"learning_rate": 4.993442008619075e-05,
"loss": 152.2921,
"step": 8
},
{
"epoch": 0.0014991274609699823,
"eval_accuracy": 0.015459481333930691,
"eval_loss": 10.50749397277832,
"eval_runtime": 140.9589,
"eval_samples_per_second": 56.002,
"eval_steps_per_second": 7.002,
"step": 8
},
{
"epoch": 0.0029982549219399646,
"grad_norm": 33.0531005859375,
"learning_rate": 4.985947161326589e-05,
"loss": 148.7237,
"step": 16
},
{
"epoch": 0.0029982549219399646,
"eval_accuracy": 0.018715353804478252,
"eval_loss": 10.448219299316406,
"eval_runtime": 215.51,
"eval_samples_per_second": 36.629,
"eval_steps_per_second": 4.58,
"step": 16
},
{
"epoch": 0.005996509843879929,
"grad_norm": 33.14303207397461,
"learning_rate": 4.970957466741615e-05,
"loss": 141.2197,
"step": 32
},
{
"epoch": 0.005996509843879929,
"eval_accuracy": 0.02386372614066984,
"eval_loss": 10.312174797058105,
"eval_runtime": 201.4605,
"eval_samples_per_second": 39.184,
"eval_steps_per_second": 4.899,
"step": 32
},
{
"epoch": 0.011993019687759858,
"grad_norm": 17.63327407836914,
"learning_rate": 4.94097807757167e-05,
"loss": 128.0666,
"step": 64
},
{
"epoch": 0.011993019687759858,
"eval_accuracy": 0.0342046782626398,
"eval_loss": 9.543290138244629,
"eval_runtime": 210.7504,
"eval_samples_per_second": 37.457,
"eval_steps_per_second": 4.683,
"step": 64
},
{
"epoch": 0.023986039375519717,
"grad_norm": 22.085046768188477,
"learning_rate": 4.881019299231778e-05,
"loss": 118.9934,
"step": 128
},
{
"epoch": 0.023986039375519717,
"eval_accuracy": 0.03304265387350131,
"eval_loss": 7.98660945892334,
"eval_runtime": 227.9954,
"eval_samples_per_second": 34.624,
"eval_steps_per_second": 4.329,
"step": 128
},
{
"epoch": 0.04797207875103943,
"grad_norm": 17.06017303466797,
"learning_rate": 4.761101742551995e-05,
"loss": 102.9587,
"step": 256
},
{
"epoch": 0.04797207875103943,
"eval_accuracy": 0.055030721081703045,
"eval_loss": 5.9021315574646,
"eval_runtime": 240.335,
"eval_samples_per_second": 32.846,
"eval_steps_per_second": 4.107,
"step": 256
},
{
"epoch": 0.09594415750207887,
"grad_norm": 68.73705291748047,
"learning_rate": 4.52126662919243e-05,
"loss": 93.4498,
"step": 512
},
{
"epoch": 0.09594415750207887,
"eval_accuracy": 0.18273613650666046,
"eval_loss": 4.6600799560546875,
"eval_runtime": 236.8236,
"eval_samples_per_second": 33.333,
"eval_steps_per_second": 4.168,
"step": 512
}
],
"logging_steps": 99999999,
"max_steps": 5337,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 99999999,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 3.4248069218304e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}