Detect GigaChat3-10-A1.8B as deepseek lite

[ubergarm]

Hardcodes checking number of layers to detect if a model is the lite version of deepseek.

Tested with bf16 and q8_0 version of GigaChat3-10B-A1.8B and discussed realizing it was a lite version similar to DeepSeek-V2-Lite. That model had 27 layers, but GigaChat3 has 26 and that is used to detect the lite variant as discussed here: https://huggingface.co/ai-sage/GigaChat3-10B-A1.8B/discussions/1#691fb161ac024c8eb626ab36

I'd like if anyone else could test. I'll update after testing perplexity to make sure the value looks sane. I haven't uploaded a gguf yet as the template has a parse error and wanted to get it updated before baking it in. That is discussed here: https://huggingface.co/ai-sage/GigaChat3-702B-A36B-preview-bf16/discussions/1

[ubergarm]

I added a test Q8_0 gguf here: https://huggingface.co/ubergarm/GigaChat3-10B-A1.8B-GGUF/tree/main

$ export model=/mnt/data/models/ubergarm/GigaChat3-10B-A1.8B-GGUF/GigaChat3-10B-A1.8B-Q8_0.gguf
$ ./build/bin/llama-server \
    --model "$model"\
    --alias ubergarm/GigaChat3-10B-A1.8B-GGUF \
    --ctx-size 32768 \
    --parallel 1 \
    --threads 8 \
    --host 127.0.0.1 \
    --port 8080 \
    --jinja

llama_model_loader: - type  f32:  129 tensors
llama_model_loader: - type q8_0:  285 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q8_0
print_info: file size   = 10.57 GiB (8.51 BPW)

[ubergarm]

Oops i need to get rid of an accidently commit file, will force push to fix.

[ubergarm]

Perplexity seems reasonable:

• ==> logs/perplexity-GigaChat3-10B-A1.8B-BF16.log <==
  • Final estimate: PPL = 6.7302 +/- 0.04230
• ==> logs/perplexity-GigaChat3-10B-A1.8B-Q8_0.log <==
  • Final estimate: PPL = 6.7265 +/- 0.04226

Always a bit funky when quants have lower perplexity than the original bf16... though it happens sometimes and in this case the values are very similar within the noise.

[CISC]

Can you add a comment describing which models we are detecting in case this needs to be fine-tuned in the future?

[CISC]

Thank you!