DeepSeek-R1-Distill-Qwen-7B-GGUF / scores /deepseek-r1-distill-qwen-7b-q4_k_m-bartowski.arc
eaddario's picture
Regenerate Perplexity, KLD, ARC, HellaSwag, MMLU, Truthful QA and WinoGrande scores
0f51238 verified
raw
history blame
834 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 30 key-value pairs and 339 tensors from ./DeepSeek-R1-Distill-Qwen-7B-Q4_K_M-bartowski.gguf (version GGUF V3 (latest))
Final result: 50.9358 +/- 1.8291
Random chance: 25.0335 +/- 1.5850
llama_perf_context_print: load time = 1777.07 ms
llama_perf_context_print: prompt eval time = 156585.88 ms / 36626 tokens ( 4.28 ms per token, 233.90 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 = 158452.54 ms / 36627 tokens
ggml_metal_free: deallocating