Int8OPTAttention Class Reference

Public Member Functions
	Int8OPTAttention (std::string param_path, const struct model_config config, BMM_S8T_S8N_F32T &qk_bmm, BMM_S8T_S8N_S8T &pv_bmm, W8A8B8O8Linear &k_proj, W8A8B8O8Linear &v_proj, W8A8B8O8Linear &q_proj, W8A8BFP32OFP32Linear &out_proj)
struct Int8OPTAttention_output	forward (const struct Int8OPTAttention_input &input)

Static Public Member Functions
static void	initialized_memory (const struct model_config config)

---

The documentation for this class was generated from the following files:

• llm/include/nn_modules/Int8OPTAttention.h
• llm/src/nn_modules/Int8OPTAttention.cc