GoLLeM-v6-250M / config.json
Maggio33's picture
GoLLeM v6: weights (safetensors), config, tokenizer, model code, card
bf9f6d6 verified
Raw History Blame Contribute Delete
354 Bytes
{
"block": 1024,
"ffn": "swiglu",
"ffn_mult": 2.667,
"logit_cap": 0.0,
"n_embd": 960,
"n_head": 15,
"n_layer": 20,
"norm": "rmsnorm",
"norm_eps": 1e-06,
"pos": "rope",
"qk_norm": true,
"rope_theta": 100000.0,
"tied_weights": {
"tok.weight": "head.weight"
},
"value_residual": true,
"vocab": 32768
}