nanotron
/

llama3-8b-infini-attention

Model card Files Files and versions Community

llama3-8b-infini-attention / checkpoint_metadata.json

neuralink's picture

neuralink HF staff

add ckp

0610800 4 months ago

history blame contribute delete

130 Bytes

	{
	"dp": 6,
	"metas": {
	"consumed_train_samples": 1920000,
	"last_train_step": 20000
	},
	"tp": 4,
	"version": "1.2"
	}