52 lines
979 B
JSON
52 lines
979 B
JSON
{
|
|
"wandb_run": "run-20260630_063402-76k27fnx",
|
|
"wandb_output_log": "/mnt/mole/SDPO/L2T/wandb/run-20260630_063402-76k27fnx/files/output.log",
|
|
"metric_key": "val-aux/tooluse/reward/mean@16",
|
|
"evals": [
|
|
{
|
|
"step": 10,
|
|
"mean16": 0.5992647058823529
|
|
},
|
|
{
|
|
"step": 20,
|
|
"mean16": 0.5606617647058824
|
|
},
|
|
{
|
|
"step": 30,
|
|
"mean16": 0.6029411764705882
|
|
},
|
|
{
|
|
"step": 40,
|
|
"mean16": 0.5882352941176471
|
|
},
|
|
{
|
|
"step": 50,
|
|
"mean16": 0.5955882352941176
|
|
},
|
|
{
|
|
"step": 60,
|
|
"mean16": 0.609375
|
|
},
|
|
{
|
|
"step": 70,
|
|
"mean16": 0.6341911764705882
|
|
},
|
|
{
|
|
"step": 80,
|
|
"mean16": 0.6121323529411765
|
|
},
|
|
{
|
|
"step": 90,
|
|
"mean16": 0.5845588235294118
|
|
},
|
|
{
|
|
"step": 100,
|
|
"mean16": 0.6011029411764706
|
|
}
|
|
],
|
|
"best_step": 70,
|
|
"best_mean16": 0.6341911764705882,
|
|
"final_step": 100,
|
|
"final_mean16": 0.6011029411764706
|
|
}
|