|
TinyChatEngine
|
Public Member Functions | |
| Fp32llamaDecoderLayer (std::string param_path, const struct model_config config, int layer_idx) | |
| struct Fp32llamaDecoderLayer_output | forward (const struct Fp32llamaDecoderLayer_input &input) |
Public Attributes | |
| int | embed_dim |
| int | num_attention_heads |
| int | hidden_dim |
| int | layer_idx |
| float | rms_norm_eps |
| LlamaRMSNorm | input_layernorm |
| LlamaRMSNorm | post_attention_layernorm |
| Linear_FP | gate_proj |
| Linear_FP | down_proj |
| Linear_FP | up_proj |
| Fp32llamaAttention | attn |
| std::string | profile_name = "Fp32llamaDecoderLayer" |