dctanner
/

sablo-pebble-mistral-dpo-lora-oasst2_dpo_pairs_en

alignment-handbook

Generated from Trainer

Model card Files Files and versions Metrics Training metrics Community

sablo-pebble-mistral-dpo-lora-oasst2_dpo_pairs_en / train_results.json

dctanner's picture

Model save

82029ca verified 4 months ago

raw history blame contribute delete

No virus

193 Bytes

	{
	"epoch": 1.0,
	"train_loss": 0.6752520862378573,
	"train_runtime": 2732.0836,
	"train_samples": 4860,
	"train_samples_per_second": 1.779,
	"train_steps_per_second": 0.028
	}