TheBloke
/

Llama-2-13B-Chat-Dutch-GPTQ

Text Generation

generated_from_trainer

text-generation-inference

4-bit precision

Model card Files Files and versions Community

Llama-2-13B-Chat-Dutch-GPTQ / eval_results.json

TheBloke's picture

GPTQ model commit

6a68220 8 months ago

raw history blame

No virus

224 Bytes

	{
	"epoch": 2.0,
	"eval_loss": 1.0847766399383545,
	"eval_runtime": 298.8617,
	"eval_samples": 630,
	"eval_samples_per_second": 2.108,
	"eval_steps_per_second": 0.264,
	"perplexity": 2.9587788718964534
	}