DeepSeek-R1-Distill-Qwen-7B-GGUF / scores /deepseek-r1-distill-qwen-7b-q3_k_m.wng
eaddario's picture
Regenerate Perplexity, KLD, ARC, HellaSwag, MMLU, Truthful QA and WinoGrande scores
0f51238 verified
raw
history blame
810 Bytes
build: 4998 (d41f0150) with Apple clang version 16.0.0 (clang-1600.0.26.6) for arm64-apple-darwin24.3.0
llama_model_load_from_file_impl: using device Metal (Apple M3 Pro) - 27647 MiB free
llama_model_loader: loaded meta data with 31 key-value pairs and 339 tensors from ./DeepSeek-R1-Distill-Qwen-7B-Q3_K_M.gguf (version GGUF V3 (latest))
Final Winogrande score(750 tasks): 61.2000 +/- 1.7805
llama_perf_context_print: load time = 240.77 ms
llama_perf_context_print: prompt eval time = 90380.84 ms / 22286 tokens ( 4.06 ms per token, 246.58 tokens per second)
llama_perf_context_print: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second)
llama_perf_context_print: total time = 91416.04 ms / 22287 tokens
ggml_metal_free: deallocating