diff options
| author | Andrew Canis <andrew.canis@gmail.com> | 2024-03-15 16:41:22 -0400 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-03-15 22:41:22 +0200 |
| commit | 12247f4c69a173b9482f68aaa174ec37fc909ccf (patch) | |
| tree | 1c580de91d5d0676e146bb45b9197d88aeb226fd /gguf-py | |
| parent | 4e9a7f7f7fb6acbddd1462909c8d696e38edbfcc (diff) | |
llama : add Command-R support (#6033)
Information about the Command-R 35B model (128k context) can be found at:
https://huggingface.co/CohereForAI/c4ai-command-r-v01
Based on the llama2 model with a few changes:
1) New hyper parameter to scale output logits (logit_scale)
2) Uses LayerNorm instead of RMSNorm
3) Transfomer layers have a single shared LayerNorm that feeds into both the
self-attention and FFN layers in parallel. There is no post-attention LayerNorm.
4) No support for Rotary Position Embeddings (RoPE) scaling
5) No biases used
Find GGUF files here:
https://huggingface.co/andrewcanis/c4ai-command-r-v01-GGUF
To convert model to GGUF format yourself:
1) Download Command-R Hugging Face safetensors:
git lfs install
git clone https://huggingface.co/CohereForAI/c4ai-command-r-v01
2) Run:
python3 convert-hf-to-gguf.py --outtype f16 ./c4ai-command-r-v01
Diffstat (limited to 'gguf-py')
| -rw-r--r-- | gguf-py/gguf/constants.py | 15 | ||||
| -rw-r--r-- | gguf-py/gguf/gguf_writer.py | 3 |
2 files changed, 18 insertions, 0 deletions
diff --git a/gguf-py/gguf/constants.py b/gguf-py/gguf/constants.py index 458a641d..4a4facb0 100644 --- a/gguf-py/gguf/constants.py +++ b/gguf-py/gguf/constants.py @@ -42,6 +42,7 @@ class Keys: EXPERT_COUNT = "{arch}.expert_count" EXPERT_USED_COUNT = "{arch}.expert_used_count" POOLING_TYPE = "{arch}.pooling_type" + LOGIT_SCALE = "{arch}.logit_scale" class Attention: HEAD_COUNT = "{arch}.attention.head_count" @@ -121,6 +122,7 @@ class MODEL_ARCH(IntEnum): GEMMA = auto() STARCODER2 = auto() MAMBA = auto() + COMMAND_R = auto() class MODEL_TENSOR(IntEnum): @@ -187,6 +189,7 @@ MODEL_ARCH_NAMES: dict[MODEL_ARCH, str] = { MODEL_ARCH.GEMMA: "gemma", MODEL_ARCH.STARCODER2: "starcoder2", MODEL_ARCH.MAMBA: "mamba", + MODEL_ARCH.COMMAND_R: "command-r", } TENSOR_NAMES: dict[MODEL_TENSOR, str] = { @@ -579,6 +582,18 @@ MODEL_TENSORS: dict[MODEL_ARCH, list[MODEL_TENSOR]] = { MODEL_TENSOR.SSM_D, MODEL_TENSOR.SSM_OUT, ], + MODEL_ARCH.COMMAND_R: [ + MODEL_TENSOR.TOKEN_EMBD, + MODEL_TENSOR.OUTPUT_NORM, + MODEL_TENSOR.ATTN_NORM, + MODEL_TENSOR.ATTN_Q, + MODEL_TENSOR.ATTN_K, + MODEL_TENSOR.ATTN_V, + MODEL_TENSOR.ATTN_OUT, + MODEL_TENSOR.FFN_GATE, + MODEL_TENSOR.FFN_DOWN, + MODEL_TENSOR.FFN_UP, + ], # TODO } diff --git a/gguf-py/gguf/gguf_writer.py b/gguf-py/gguf/gguf_writer.py index 1967b633..2ae6c814 100644 --- a/gguf-py/gguf/gguf_writer.py +++ b/gguf-py/gguf/gguf_writer.py @@ -361,6 +361,9 @@ class GGUFWriter: def add_clamp_kqv(self, value: float) -> None: self.add_float32(Keys.Attention.CLAMP_KQV.format(arch=self.arch), value) + def add_logit_scale(self, value: float) -> None: + self.add_float32(Keys.LLM.LOGIT_SCALE.format(arch=self.arch), value) + def add_expert_count(self, count: int) -> None: self.add_uint32(Keys.LLM.EXPERT_COUNT.format(arch=self.arch), count) |