From f6793491b5af6da75edad34d6f503ef86d31b09f Mon Sep 17 00:00:00 2001 From: "Nam D. Tran" <42194884+namtranase@users.noreply.github.com> Date: Wed, 27 Dec 2023 22:39:45 +0700 Subject: llama : add AWQ for llama, llama2, mpt, and mistral models (#4593) MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit * update: awq support llama-7b model * update: change order * update: benchmark results for llama2-7b * update: mistral 7b v1 benchmark * update: support 4 models * fix: Readme * update: ready for PR * update: readme * fix: readme * update: change order import * black * format code * update: work for bot mpt and awqmpt * update: readme * Rename to llm_build_ffn_mpt_awq * Formatted other files * Fixed params count * fix: remove code * update: more detail for mpt * fix: readme * fix: readme * update: change folder architecture * fix: common.cpp * fix: readme * fix: remove ggml_repeat * update: cicd * update: cicd * uppdate: remove use_awq arg * update: readme * llama : adapt plamo to new ffn ggml-ci --------- Co-authored-by: Trần Đức Nam Co-authored-by: Le Hoang Anh Co-authored-by: Georgi Gerganov --- gguf-py/gguf/tensor_mapping.py | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'gguf-py/gguf/tensor_mapping.py') diff --git a/gguf-py/gguf/tensor_mapping.py b/gguf-py/gguf/tensor_mapping.py index 446c6b68..0b8f7041 100644 --- a/gguf-py/gguf/tensor_mapping.py +++ b/gguf-py/gguf/tensor_mapping.py @@ -188,6 +188,11 @@ class TensorNameMap: "model.layers.{bid}.block_sparse_moe.experts.{xid}.w3", # mixtral ), + # AWQ-activation gate + MODEL_TENSOR.FFN_ACT: ( + "transformer.blocks.{bid}.ffn.act", # mpt + ), + # Feed-forward gate MODEL_TENSOR.FFN_GATE: ( "model.layers.{bid}.mlp.gate_proj", # llama-hf refact -- cgit v1.2.3