DeepSeek-R1-Distill-Llama-8B-GGUF / scores /DeepSeek-R1-Distill-Llama-8B-q3_k_s.tqa
eaddario's picture
Generate Perplexity, KLD, ARC, HellaSwag, MMLU, Truthful QA and WinoGrande scores
4a6a053 verified
raw
history blame
795 Bytes
build: 4898 (8a827f88) 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 33 key-value pairs and 292 tensors from ./DeepSeek-R1-Distill-Llama-8B-Q3_K_S.gguf (version GGUF V3 (latest))
Final result: 34.5679 +/- 2.6462
Random chance: 28.5714 +/- 2.5136
llama_perf_context_print: load time = 309.73 ms
llama_perf_context_print: prompt eval time = 84156.81 ms / 17573 tokens ( 4.79 ms per token, 208.81 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 = 85450.77 ms / 17574 tokens