max-rl
/

maze_17x17_sft_bs256_checkpoints

Reinforcement Learning

Model card Files Files and versions

maze_17x17_sft_bs256_checkpoints / random /init_model /tokenizer_config.json

guanning's picture

add bs256_random run (init_model + ckpt-250..ckpt-4250)

d7b1a64 verified about 1 month ago

history blame contribute delete

173 Bytes

	{
	"bos_token": "<bos>",
	"eos_token": "<eos>",
	"unk_token": "<unk>",
	"pad_token": "<pad>",
	"tokenizer_class": "PreTrainedTokenizerFast",
	"model_max_length": 512
	}