warning: no usable GPU found, --gpu-layers option will be ignored warning: one possible reason is that llama.cpp was compiled without GPU support warning: consult docs/build.md for compilation instructions 0.00.350.172 W load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect 0.01.333.570 I cmn init: llama threadpool init, n_threads = 64 0.01.410.456 I 0.01.410.625 I system_info: n_threads = 64 (n_threads_batch = 64) / 80 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX_VNNI = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | AVX512 = 1 | AVX512_VBMI = 1 | AVX512_VNNI = 1 | AVX512_BF16 = 1 | AMX_INT8 = 1 | LLAMAFILE = 1 | OPENMP = 1 | REPACK = 1 | 0.01.410.641 I perplexity: tokenizing the input .. 0.02.026.275 I perplexity: tokenization took 615.627 ms 0.02.026.496 I perplexity: calculating perplexity over 64 chunks, n_ctx=512, batch_size=2048, n_seq=4 0.07.853.203 I perplexity: 5.83 seconds per pass - ETA 1.55 minutes [1]8.1805,[2]12.6373,[3]11.8638,[4]12.3617,[5]12.6663,[6]13.1201,[7]13.8095,[8]14.2267,[9]14.9397,[10]15.5506,[11]15.7619,[12]16.0234,[13]16.0860,[14]15.7798,[15]15.4154,[16]15.8763,[17]14.5611,[18]14.6866,[19]14.8310,[20]14.5036,[21]14.3380,[22]14.1237,[23]13.6626,[24]13.2579,[25]12.8837,[26]12.7161,[27]12.0963,[28]11.9504,[29]11.9826,[30]11.7738,[31]11.9129,[32]11.9238,[33]12.0048,[34]12.2665,[35]12.3705,[36]12.5638,[37]12.6779,[38]12.8489,[39]12.7929,[40]12.7321,[41]12.7124,[42]12.6202,[43]12.7620,[44]12.7058,[45]12.6333,[46]12.8092,[47]12.7222,[48]12.9243,[49]12.9285,[50]12.9425,[51]13.0944,[52]13.1409,[53]13.1278,[54]13.1657,[55]13.2771,[56]13.3026,[57]13.3324,[58]13.3268,[59]13.4020,[60]13.4787,[61]13.5656,[62]13.6147,[63]13.6801,[64]13.8127, 1.21.268.982 I Final estimate: PPL = 13.8127 +/- 0.34818