diff options
| author | Guoteng <32697156+SolenoidWGT@users.noreply.github.com> | 2024-02-01 17:19:51 +0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-02-01 11:19:51 +0200 |
| commit | ce32060198b7e2d6a13a9b8e1e1369e3c295ae2a (patch) | |
| tree | 546b5a7a327d3f3f1370549331915c4d6373d51d /gguf-py/gguf/constants.py | |
| parent | 1cfb5372cf5707c8ec6dde7c874f4a44a6c4c915 (diff) | |
llama : support InternLM2 (#5184)
* support InternLM2 inference
* add add_space_prefix KV pair
Diffstat (limited to 'gguf-py/gguf/constants.py')
| -rw-r--r-- | gguf-py/gguf/constants.py | 18 |
1 files changed, 18 insertions, 0 deletions
diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py index f5c933a4..ed8e26f8 100644 --- a/gguf-py/gguf/constants.py +++ b/gguf-py/gguf/constants.py @@ -72,6 +72,7 @@ class Keys: PAD_ID = "tokenizer.ggml.padding_token_id" ADD_BOS = "tokenizer.ggml.add_bos_token" ADD_EOS = "tokenizer.ggml.add_eos_token" + ADD_PREFIX = "tokenizer.ggml.add_space_prefix" HF_JSON = "tokenizer.huggingface.json" RWKV = "tokenizer.rwkv.world" CHAT_TEMPLATE = "tokenizer.chat_template" @@ -102,6 +103,7 @@ class MODEL_ARCH(IntEnum): PLAMO = auto() CODESHELL = auto() ORION = auto() + INTERNLM2 = auto() class MODEL_TENSOR(IntEnum): @@ -153,6 +155,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = { MODEL_ARCH.PLAMO: "plamo", MODEL_ARCH.CODESHELL: "codeshell", MODEL_ARCH.ORION: "orion", + MODEL_ARCH.INTERNLM2: "internlm2", } TENSOR_NAMES: dict[MODEL_TENSOR, str] = { @@ -446,6 +449,21 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = { MODEL_TENSOR.FFN_DOWN, MODEL_TENSOR.FFN_UP, ], + MODEL_ARCH.INTERNLM2: [ + MODEL_TENSOR.TOKEN_EMBD, + MODEL_TENSOR.OUTPUT_NORM, + MODEL_TENSOR.OUTPUT, + MODEL_TENSOR.ATTN_NORM, + MODEL_TENSOR.ATTN_Q, + MODEL_TENSOR.ATTN_K, + MODEL_TENSOR.ATTN_V, + MODEL_TENSOR.ATTN_OUT, + MODEL_TENSOR.ATTN_ROT_EMBD, + MODEL_TENSOR.FFN_NORM, + MODEL_TENSOR.FFN_GATE, + MODEL_TENSOR.FFN_DOWN, + MODEL_TENSOR.FFN_UP, + ], # TODO } |