chansung
/

Qwen3-4B-CCRL-CUR-VAR-ASCE-NORMAL-1E

Text Generation

Generated from Trainer

text-generation-inference

Model card Files Files and versions

Qwen3-4B-CCRL-CUR-VAR-ASCE-NORMAL-1E / all_results.json

chansung's picture

Model save

274927f verified 3 months ago

201 Bytes

	{
	"total_flos": 0.0,
	"train_loss": 0.08443626471047583,
	"train_runtime": 92210.0342,
	"train_samples": 15506,
	"train_samples_per_second": 0.168,
	"train_steps_per_second": 0.003
	}