seed_oss

seed_oss

Seed-OSS (SeedOssForCausalLM).

Llama's tree and Llama's block: model.{embed_tokens, layers[i].{input_layernorm, self_attn, post_attention_layernorm, mlp}, norm} and lm_head, the residual added in the block, attention through the shared eager forward. The attention output and the MLP output each pass a residual dropout inside their module, which is the identity in eval.

RENAME module-attribute

RENAME = {'model.embed_tokens': 'embed_tokens', 'model.layers': 'layers', 'model.norm': 'norm'}

ENVOYS module-attribute

ENVOYS = {SeedOssDecoderLayer: Layer, SeedOssAttention: Attention, SeedOssMLP: Mlp}