Alpha55 / R1 /experiment_config.json
MSkilll's picture
The final model (R1) is now out.
b776003 verified
Raw History Blame Contribute Delete
199 Bytes
{
"model_type": "encoder-decoder-transformer",
"d_model": 256,
"layers": {
"encoder": 3,
"decoder": 3
},
"vocab_size": 32000,
"parameters": 30137600
}