From dacfcebd6022175848e978f82811a244f1033038 Mon Sep 17 00:00:00 2001 From: Victor Nogueira Date: Thu, 23 May 2024 15:12:43 +0300 Subject: [PATCH 1/4] readme : add GPT-NeoX + Pythia to the list of supported models (#7491) --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index f4088c05e..ccd12e211 100644 --- a/README.md +++ b/README.md @@ -127,6 +127,7 @@ Typically finetunes of the base models below are supported as well. - [x] [SEA-LION](https://huggingface.co/models?search=sea-lion) - [x] [GritLM-7B](https://huggingface.co/GritLM/GritLM-7B) + [GritLM-8x7B](https://huggingface.co/GritLM/GritLM-8x7B) - [x] [OLMo](https://allenai.org/olmo) +- [x] [GPT-NeoX](https://github.com/EleutherAI/gpt-neox) + [Pythia](https://github.com/EleutherAI/pythia) (instructions for supporting more models: [HOWTO-add-model.md](./docs/HOWTO-add-model.md)) From 55ac3b7aeaf52f19786ed96e885d89521fc0f6c8 Mon Sep 17 00:00:00 2001 From: Georgi Gerganov Date: Thu, 23 May 2024 15:28:14 +0300 Subject: [PATCH 2/4] ci : use Pythia models instead of OpenLlama (#7470) * ci : start using Pythia models over OpenLlama ggml-ci * ci : disable q2_k ppl tests * ci : use convert-hf-to-gguf.py * ci : update gg_get_model * ci : fix convert outfile name ggml-ci * llama : gptneox arch use F32 attn prec ggml-ci --- ci/run.sh | 424 +++++++++++++++++++++++++++++++++++------------------- llama.cpp | 4 +- 2 files changed, 282 insertions(+), 146 deletions(-) diff --git a/ci/run.sh b/ci/run.sh index 79dcd0772..940299025 100755 --- a/ci/run.sh +++ b/ci/run.sh @@ -202,12 +202,15 @@ function gg_sum_test_scripts_release { } function gg_get_model { - local gguf_3b="$MNT/models/open-llama/3B-v2/ggml-model-f16.gguf" - local gguf_7b="$MNT/models/open-llama/7B-v2/ggml-model-f16.gguf" - if [[ -s $gguf_3b ]]; then - echo -n "$gguf_3b" - elif [[ -s $gguf_7b ]]; then - echo -n "$gguf_7b" + local gguf_0="$MNT/models/pythia/1.4B/ggml-model-f16.gguf" + local gguf_1="$MNT/models/pythia/2.8B/ggml-model-f16.gguf" + local gguf_2="$MNT/models/open-llama/7B-v2/ggml-model-f16.gguf" + if [[ -s $gguf_0 ]]; then + echo -n "$gguf_0" + elif [[ -s $gguf_1 ]]; then + echo -n "$gguf_1" + elif [[ -s $gguf_2 ]]; then + echo -n "$gguf_2" else echo >&2 "No model found. Can't run gg_run_ctest_with_model." exit 1 @@ -256,139 +259,6 @@ function gg_sum_ctest_with_model_release { gg_printf '```\n' } -# open_llama_3b_v2 - -function gg_run_open_llama_3b_v2 { - cd ${SRC} - - gg_wget models-mnt/open-llama/3B-v2/ https://huggingface.co/openlm-research/open_llama_3b_v2/raw/main/config.json - gg_wget models-mnt/open-llama/3B-v2/ https://huggingface.co/openlm-research/open_llama_3b_v2/resolve/main/tokenizer.model - gg_wget models-mnt/open-llama/3B-v2/ https://huggingface.co/openlm-research/open_llama_3b_v2/raw/main/tokenizer_config.json - gg_wget models-mnt/open-llama/3B-v2/ https://huggingface.co/openlm-research/open_llama_3b_v2/raw/main/special_tokens_map.json - gg_wget models-mnt/open-llama/3B-v2/ https://huggingface.co/openlm-research/open_llama_3b_v2/resolve/main/pytorch_model.bin - gg_wget models-mnt/open-llama/3B-v2/ https://huggingface.co/openlm-research/open_llama_3b_v2/raw/main/generation_config.json - - gg_wget models-mnt/wikitext/ https://huggingface.co/datasets/ggml-org/ci/resolve/main/wikitext-2-raw-v1.zip - unzip -o models-mnt/wikitext/wikitext-2-raw-v1.zip -d models-mnt/wikitext/ - head -n 60 models-mnt/wikitext/wikitext-2-raw/wiki.test.raw > models-mnt/wikitext/wikitext-2-raw/wiki.test-60.raw - - path_models="../models-mnt/open-llama/3B-v2" - path_wiki="../models-mnt/wikitext/wikitext-2-raw" - - rm -rf build-ci-release && mkdir build-ci-release && cd build-ci-release - - set -e - - (time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} -DLLAMA_QKK_64=1 .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log - (time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log - - python3 ../convert.py ${path_models} - - model_f16="${path_models}/ggml-model-f16.gguf" - model_q8_0="${path_models}/ggml-model-q8_0.gguf" - model_q4_0="${path_models}/ggml-model-q4_0.gguf" - model_q4_1="${path_models}/ggml-model-q4_1.gguf" - model_q5_0="${path_models}/ggml-model-q5_0.gguf" - model_q5_1="${path_models}/ggml-model-q5_1.gguf" - model_q2_k="${path_models}/ggml-model-q2_k.gguf" - model_q3_k="${path_models}/ggml-model-q3_k.gguf" - model_q4_k="${path_models}/ggml-model-q4_k.gguf" - model_q5_k="${path_models}/ggml-model-q5_k.gguf" - model_q6_k="${path_models}/ggml-model-q6_k.gguf" - - wiki_test_60="${path_wiki}/wiki.test-60.raw" - - ./bin/quantize ${model_f16} ${model_q8_0} q8_0 - ./bin/quantize ${model_f16} ${model_q4_0} q4_0 - ./bin/quantize ${model_f16} ${model_q4_1} q4_1 - ./bin/quantize ${model_f16} ${model_q5_0} q5_0 - ./bin/quantize ${model_f16} ${model_q5_1} q5_1 - ./bin/quantize ${model_f16} ${model_q2_k} q2_k - ./bin/quantize ${model_f16} ${model_q3_k} q3_k - ./bin/quantize ${model_f16} ${model_q4_k} q4_k - ./bin/quantize ${model_f16} ${model_q5_k} q5_k - ./bin/quantize ${model_f16} ${model_q6_k} q6_k - - (time ./bin/main --model ${model_f16} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log - (time ./bin/main --model ${model_q8_0} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log - (time ./bin/main --model ${model_q4_0} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log - (time ./bin/main --model ${model_q4_1} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log - (time ./bin/main --model ${model_q5_0} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log - (time ./bin/main --model ${model_q5_1} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log - (time ./bin/main --model ${model_q2_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log - (time ./bin/main --model ${model_q3_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log - (time ./bin/main --model ${model_q4_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log - (time ./bin/main --model ${model_q5_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log - (time ./bin/main --model ${model_q6_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log - - (time ./bin/perplexity --model ${model_f16} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log - (time ./bin/perplexity --model ${model_q8_0} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log - (time ./bin/perplexity --model ${model_q4_0} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log - (time ./bin/perplexity --model ${model_q4_1} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log - (time ./bin/perplexity --model ${model_q5_0} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log - (time ./bin/perplexity --model ${model_q5_1} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log - (time ./bin/perplexity --model ${model_q2_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log - (time ./bin/perplexity --model ${model_q3_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log - (time ./bin/perplexity --model ${model_q4_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log - (time ./bin/perplexity --model ${model_q5_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log - (time ./bin/perplexity --model ${model_q6_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log - - (time ./bin/imatrix --model ${model_f16} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-imatrix.log - - (time ./bin/save-load-state --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log - (time ./bin/save-load-state -fa --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log - - function check_ppl { - qnt="$1" - ppl=$(echo "$2" | grep -oE "[0-9]+\.[0-9]+" | tail -n 1) - - if [ $(echo "$ppl > 20.0" | bc) -eq 1 ]; then - printf ' - %s @ %s (FAIL: ppl > 20.0)\n' "$qnt" "$ppl" - return 20 - fi - - printf ' - %s @ %s OK\n' "$qnt" "$ppl" - return 0 - } - - check_ppl "f16" "$(cat $OUT/${ci}-tg-f16.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q8_0" "$(cat $OUT/${ci}-tg-q8_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q4_0" "$(cat $OUT/${ci}-tg-q4_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q4_1" "$(cat $OUT/${ci}-tg-q4_1.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q5_0" "$(cat $OUT/${ci}-tg-q5_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q5_1" "$(cat $OUT/${ci}-tg-q5_1.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q2_k" "$(cat $OUT/${ci}-tg-q2_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q3_k" "$(cat $OUT/${ci}-tg-q3_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q4_k" "$(cat $OUT/${ci}-tg-q4_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q5_k" "$(cat $OUT/${ci}-tg-q5_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - check_ppl "q6_k" "$(cat $OUT/${ci}-tg-q6_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log - - cat $OUT/${ci}-imatrix.log | grep "Final" >> $OUT/${ci}-imatrix-sum.log - - set +e -} - -function gg_sum_open_llama_3b_v2 { - gg_printf '### %s\n\n' "${ci}" - - gg_printf 'OpenLLaMA 3B-v2:\n' - gg_printf '- status: %s\n' "$(cat $OUT/${ci}.exit)" - gg_printf '- perplexity:\n%s\n' "$(cat $OUT/${ci}-ppl.log)" - gg_printf '- imatrix:\n```\n%s\n```\n' "$(cat $OUT/${ci}-imatrix-sum.log)" - gg_printf '- f16: \n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-f16.log)" - gg_printf '- q8_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q8_0.log)" - gg_printf '- q4_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_0.log)" - gg_printf '- q4_1:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_1.log)" - gg_printf '- q5_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_0.log)" - gg_printf '- q5_1:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_1.log)" - gg_printf '- q2_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q2_k.log)" - gg_printf '- q3_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q3_k.log)" - gg_printf '- q4_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_k.log)" - gg_printf '- q5_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_k.log)" - gg_printf '- q6_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q6_k.log)" - gg_printf '- save-load-state: \n```\n%s\n```\n' "$(cat $OUT/${ci}-save-load-state.log)" -} - # open_llama_7b_v2 # requires: GG_BUILD_CUDA @@ -417,7 +287,7 @@ function gg_run_open_llama_7b_v2 { (time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} -DLLAMA_CUDA=1 .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log (time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log - python3 ../convert.py ${path_models} + python3 ../convert.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf model_f16="${path_models}/ggml-model-f16.gguf" model_q8_0="${path_models}/ggml-model-q8_0.gguf" @@ -526,6 +396,272 @@ function gg_sum_open_llama_7b_v2 { gg_printf '- save-load-state: \n```\n%s\n```\n' "$(cat $OUT/${ci}-save-load-state.log)" } +# pythia_1.4b + +function gg_run_pythia_1_4b { + cd ${SRC} + + gg_wget models-mnt/pythia/1.4B/ https://huggingface.co/EleutherAI/pythia-1.4b/raw/main/config.json + gg_wget models-mnt/pythia/1.4B/ https://huggingface.co/EleutherAI/pythia-1.4b/raw/main/tokenizer.json + gg_wget models-mnt/pythia/1.4B/ https://huggingface.co/EleutherAI/pythia-1.4b/raw/main/tokenizer_config.json + gg_wget models-mnt/pythia/1.4B/ https://huggingface.co/EleutherAI/pythia-1.4b/raw/main/special_tokens_map.json + gg_wget models-mnt/pythia/1.4B/ https://huggingface.co/EleutherAI/pythia-1.4b/resolve/main/pytorch_model.bin + + gg_wget models-mnt/wikitext/ https://huggingface.co/datasets/ggml-org/ci/resolve/main/wikitext-2-raw-v1.zip + unzip -o models-mnt/wikitext/wikitext-2-raw-v1.zip -d models-mnt/wikitext/ + head -n 60 models-mnt/wikitext/wikitext-2-raw/wiki.test.raw > models-mnt/wikitext/wikitext-2-raw/wiki.test-60.raw + + path_models="../models-mnt/pythia/1.4B" + path_wiki="../models-mnt/wikitext/wikitext-2-raw" + + rm -rf build-ci-release && mkdir build-ci-release && cd build-ci-release + + set -e + + (time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log + (time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log + + python3 ../convert-hf-to-gguf.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf + + model_f16="${path_models}/ggml-model-f16.gguf" + model_q8_0="${path_models}/ggml-model-q8_0.gguf" + model_q4_0="${path_models}/ggml-model-q4_0.gguf" + model_q4_1="${path_models}/ggml-model-q4_1.gguf" + model_q5_0="${path_models}/ggml-model-q5_0.gguf" + model_q5_1="${path_models}/ggml-model-q5_1.gguf" + model_q2_k="${path_models}/ggml-model-q2_k.gguf" + model_q3_k="${path_models}/ggml-model-q3_k.gguf" + model_q4_k="${path_models}/ggml-model-q4_k.gguf" + model_q5_k="${path_models}/ggml-model-q5_k.gguf" + model_q6_k="${path_models}/ggml-model-q6_k.gguf" + + wiki_test_60="${path_wiki}/wiki.test-60.raw" + + ./bin/quantize ${model_f16} ${model_q8_0} q8_0 + ./bin/quantize ${model_f16} ${model_q4_0} q4_0 + ./bin/quantize ${model_f16} ${model_q4_1} q4_1 + ./bin/quantize ${model_f16} ${model_q5_0} q5_0 + ./bin/quantize ${model_f16} ${model_q5_1} q5_1 + ./bin/quantize ${model_f16} ${model_q2_k} q2_k + ./bin/quantize ${model_f16} ${model_q3_k} q3_k + ./bin/quantize ${model_f16} ${model_q4_k} q4_k + ./bin/quantize ${model_f16} ${model_q5_k} q5_k + ./bin/quantize ${model_f16} ${model_q6_k} q6_k + + (time ./bin/main --model ${model_f16} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log + (time ./bin/main --model ${model_q8_0} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log + (time ./bin/main --model ${model_q4_0} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log + (time ./bin/main --model ${model_q4_1} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log + (time ./bin/main --model ${model_q5_0} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log + (time ./bin/main --model ${model_q5_1} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log + (time ./bin/main --model ${model_q2_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log + (time ./bin/main --model ${model_q3_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log + (time ./bin/main --model ${model_q4_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log + (time ./bin/main --model ${model_q5_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log + (time ./bin/main --model ${model_q6_k} -s 1234 -n 64 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log + + (time ./bin/perplexity --model ${model_f16} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log + (time ./bin/perplexity --model ${model_q8_0} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log + (time ./bin/perplexity --model ${model_q4_0} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log + (time ./bin/perplexity --model ${model_q4_1} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log + (time ./bin/perplexity --model ${model_q5_0} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log + (time ./bin/perplexity --model ${model_q5_1} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log + (time ./bin/perplexity --model ${model_q2_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log + (time ./bin/perplexity --model ${model_q3_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log + (time ./bin/perplexity --model ${model_q4_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log + (time ./bin/perplexity --model ${model_q5_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log + (time ./bin/perplexity --model ${model_q6_k} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log + + (time ./bin/imatrix --model ${model_f16} -f ${wiki_test_60} -c 128 -b 128 --chunks 1 ) 2>&1 | tee -a $OUT/${ci}-imatrix.log + + (time ./bin/save-load-state --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log + (time ./bin/save-load-state -fa --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log + + function check_ppl { + qnt="$1" + ppl=$(echo "$2" | grep -oE "[0-9]+\.[0-9]+" | tail -n 1) + + if [ $(echo "$ppl > 20.0" | bc) -eq 1 ]; then + printf ' - %s @ %s (FAIL: ppl > 20.0)\n' "$qnt" "$ppl" + return 20 + fi + + printf ' - %s @ %s OK\n' "$qnt" "$ppl" + return 0 + } + + check_ppl "f16" "$(cat $OUT/${ci}-tg-f16.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q8_0" "$(cat $OUT/${ci}-tg-q8_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q4_0" "$(cat $OUT/${ci}-tg-q4_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q4_1" "$(cat $OUT/${ci}-tg-q4_1.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q5_0" "$(cat $OUT/${ci}-tg-q5_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q5_1" "$(cat $OUT/${ci}-tg-q5_1.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + #check_ppl "q2_k" "$(cat $OUT/${ci}-tg-q2_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log # note: ppl > 20.0 for this quant and model + check_ppl "q3_k" "$(cat $OUT/${ci}-tg-q3_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q4_k" "$(cat $OUT/${ci}-tg-q4_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q5_k" "$(cat $OUT/${ci}-tg-q5_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q6_k" "$(cat $OUT/${ci}-tg-q6_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + + cat $OUT/${ci}-imatrix.log | grep "Final" >> $OUT/${ci}-imatrix-sum.log + + set +e +} + +function gg_sum_pythia_1_4b { + gg_printf '### %s\n\n' "${ci}" + + gg_printf 'Pythia 1.4B:\n' + gg_printf '- status: %s\n' "$(cat $OUT/${ci}.exit)" + gg_printf '- perplexity:\n%s\n' "$(cat $OUT/${ci}-ppl.log)" + gg_printf '- imatrix:\n```\n%s\n```\n' "$(cat $OUT/${ci}-imatrix-sum.log)" + gg_printf '- f16: \n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-f16.log)" + gg_printf '- q8_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q8_0.log)" + gg_printf '- q4_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_0.log)" + gg_printf '- q4_1:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_1.log)" + gg_printf '- q5_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_0.log)" + gg_printf '- q5_1:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_1.log)" + gg_printf '- q2_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q2_k.log)" + gg_printf '- q3_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q3_k.log)" + gg_printf '- q4_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_k.log)" + gg_printf '- q5_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_k.log)" + gg_printf '- q6_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q6_k.log)" + gg_printf '- save-load-state: \n```\n%s\n```\n' "$(cat $OUT/${ci}-save-load-state.log)" +} + +# pythia_2_8b +# requires: GG_BUILD_CUDA + +function gg_run_pythia_2_8b { + cd ${SRC} + + gg_wget models-mnt/pythia/2.8B/ https://huggingface.co/EleutherAI/pythia-2.8b/raw/main/config.json + gg_wget models-mnt/pythia/2.8B/ https://huggingface.co/EleutherAI/pythia-2.8b/raw/main/tokenizer.json + gg_wget models-mnt/pythia/2.8B/ https://huggingface.co/EleutherAI/pythia-2.8b/raw/main/tokenizer_config.json + gg_wget models-mnt/pythia/2.8B/ https://huggingface.co/EleutherAI/pythia-2.8b/raw/main/special_tokens_map.json + gg_wget models-mnt/pythia/2.8B/ https://huggingface.co/EleutherAI/pythia-2.8b/resolve/main/pytorch_model.bin + + gg_wget models-mnt/wikitext/ https://huggingface.co/datasets/ggml-org/ci/resolve/main/wikitext-2-raw-v1.zip + unzip -o models-mnt/wikitext/wikitext-2-raw-v1.zip -d models-mnt/wikitext/ + + path_models="../models-mnt/pythia/2.8B" + path_wiki="../models-mnt/wikitext/wikitext-2-raw" + + rm -rf build-ci-release && mkdir build-ci-release && cd build-ci-release + + set -e + + (time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} -DLLAMA_CUDA=1 .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log + (time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log + + python3 ../convert-hf-to-gguf.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf + + model_f16="${path_models}/ggml-model-f16.gguf" + model_q8_0="${path_models}/ggml-model-q8_0.gguf" + model_q4_0="${path_models}/ggml-model-q4_0.gguf" + model_q4_1="${path_models}/ggml-model-q4_1.gguf" + model_q5_0="${path_models}/ggml-model-q5_0.gguf" + model_q5_1="${path_models}/ggml-model-q5_1.gguf" + model_q2_k="${path_models}/ggml-model-q2_k.gguf" + model_q3_k="${path_models}/ggml-model-q3_k.gguf" + model_q4_k="${path_models}/ggml-model-q4_k.gguf" + model_q5_k="${path_models}/ggml-model-q5_k.gguf" + model_q6_k="${path_models}/ggml-model-q6_k.gguf" + + wiki_test="${path_wiki}/wiki.test.raw" + + ./bin/quantize ${model_f16} ${model_q8_0} q8_0 + ./bin/quantize ${model_f16} ${model_q4_0} q4_0 + ./bin/quantize ${model_f16} ${model_q4_1} q4_1 + ./bin/quantize ${model_f16} ${model_q5_0} q5_0 + ./bin/quantize ${model_f16} ${model_q5_1} q5_1 + ./bin/quantize ${model_f16} ${model_q2_k} q2_k + ./bin/quantize ${model_f16} ${model_q3_k} q3_k + ./bin/quantize ${model_f16} ${model_q4_k} q4_k + ./bin/quantize ${model_f16} ${model_q5_k} q5_k + ./bin/quantize ${model_f16} ${model_q6_k} q6_k + + (time ./bin/main --model ${model_f16} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log + (time ./bin/main --model ${model_q8_0} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log + (time ./bin/main --model ${model_q4_0} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log + (time ./bin/main --model ${model_q4_1} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log + (time ./bin/main --model ${model_q5_0} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log + (time ./bin/main --model ${model_q5_1} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log + (time ./bin/main --model ${model_q2_k} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log + (time ./bin/main --model ${model_q3_k} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log + (time ./bin/main --model ${model_q4_k} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log + (time ./bin/main --model ${model_q5_k} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log + (time ./bin/main --model ${model_q6_k} -t 1 -ngl 999 -s 1234 -n 256 --ignore-eos -p "I believe the meaning of life is" ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log + + (time ./bin/perplexity --model ${model_f16} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-f16.log + (time ./bin/perplexity --model ${model_q8_0} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q8_0.log + (time ./bin/perplexity --model ${model_q4_0} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_0.log + (time ./bin/perplexity --model ${model_q4_1} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_1.log + (time ./bin/perplexity --model ${model_q5_0} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_0.log + (time ./bin/perplexity --model ${model_q5_1} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_1.log + (time ./bin/perplexity --model ${model_q2_k} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q2_k.log + (time ./bin/perplexity --model ${model_q3_k} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q3_k.log + (time ./bin/perplexity --model ${model_q4_k} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q4_k.log + (time ./bin/perplexity --model ${model_q5_k} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q5_k.log + (time ./bin/perplexity --model ${model_q6_k} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-tg-q6_k.log + + (time ./bin/imatrix --model ${model_f16} -f ${wiki_test} -t 1 -ngl 999 -c 2048 -b 512 --chunks 4 ) 2>&1 | tee -a $OUT/${ci}-imatrix.log + + (time ./bin/save-load-state -ngl 10 --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log + (time ./bin/save-load-state -fa -ngl 10 --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log + (time ./bin/save-load-state -ngl 99 --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log + (time ./bin/save-load-state -fa -ngl 99 --model ${model_q4_0} ) 2>&1 | tee -a $OUT/${ci}-save-load-state.log + + function check_ppl { + qnt="$1" + ppl=$(echo "$2" | grep -oE "[0-9]+\.[0-9]+" | tail -n 1) + + if [ $(echo "$ppl > 20.0" | bc) -eq 1 ]; then + printf ' - %s @ %s (FAIL: ppl > 20.0)\n' "$qnt" "$ppl" + return 20 + fi + + printf ' - %s @ %s OK\n' "$qnt" "$ppl" + return 0 + } + + check_ppl "f16" "$(cat $OUT/${ci}-tg-f16.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q8_0" "$(cat $OUT/${ci}-tg-q8_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q4_0" "$(cat $OUT/${ci}-tg-q4_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q4_1" "$(cat $OUT/${ci}-tg-q4_1.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q5_0" "$(cat $OUT/${ci}-tg-q5_0.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q5_1" "$(cat $OUT/${ci}-tg-q5_1.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + #check_ppl "q2_k" "$(cat $OUT/${ci}-tg-q2_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log # note: ppl > 20.0 for this quant and model + check_ppl "q3_k" "$(cat $OUT/${ci}-tg-q3_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q4_k" "$(cat $OUT/${ci}-tg-q4_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q5_k" "$(cat $OUT/${ci}-tg-q5_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + check_ppl "q6_k" "$(cat $OUT/${ci}-tg-q6_k.log | grep "^\[1\]")" | tee -a $OUT/${ci}-ppl.log + + cat $OUT/${ci}-imatrix.log | grep "Final" >> $OUT/${ci}-imatrix-sum.log + + set +e +} + +function gg_sum_pythia_2_8b { + gg_printf '### %s\n\n' "${ci}" + + gg_printf 'Pythia 2.8B:\n' + gg_printf '- status: %s\n' "$(cat $OUT/${ci}.exit)" + gg_printf '- perplexity:\n%s\n' "$(cat $OUT/${ci}-ppl.log)" + gg_printf '- imatrix:\n```\n%s\n```\n' "$(cat $OUT/${ci}-imatrix-sum.log)" + gg_printf '- f16: \n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-f16.log)" + gg_printf '- q8_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q8_0.log)" + gg_printf '- q4_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_0.log)" + gg_printf '- q4_1:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_1.log)" + gg_printf '- q5_0:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_0.log)" + gg_printf '- q5_1:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_1.log)" + gg_printf '- q2_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q2_k.log)" + gg_printf '- q3_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q3_k.log)" + gg_printf '- q4_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q4_k.log)" + gg_printf '- q5_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q5_k.log)" + gg_printf '- q6_k:\n```\n%s\n```\n' "$(cat $OUT/${ci}-tg-q6_k.log)" + gg_printf '- save-load-state: \n```\n%s\n```\n' "$(cat $OUT/${ci}-save-load-state.log)" +} + # bge-small function gg_run_embd_bge_small { @@ -552,7 +688,7 @@ function gg_run_embd_bge_small { (time cmake -DCMAKE_BUILD_TYPE=Release ${CMAKE_EXTRA} .. ) 2>&1 | tee -a $OUT/${ci}-cmake.log (time make -j ) 2>&1 | tee -a $OUT/${ci}-make.log - python3 ../convert-hf-to-gguf.py ${path_models} + python3 ../convert-hf-to-gguf.py ${path_models} --outfile ${path_models}/ggml-model-f16.gguf model_f16="${path_models}/ggml-model-f16.gguf" model_q8_0="${path_models}/ggml-model-q8_0.gguf" @@ -606,10 +742,10 @@ if [ -z ${GG_BUILD_LOW_PERF} ]; then if [ -z ${GG_BUILD_VRAM_GB} ] || [ ${GG_BUILD_VRAM_GB} -ge 8 ]; then if [ -z ${GG_BUILD_CUDA} ]; then - #test $ret -eq 0 && gg_run open_llama_3b_v2 - date # dummy + test $ret -eq 0 && gg_run pythia_1_4b else - test $ret -eq 0 && gg_run open_llama_7b_v2 + test $ret -eq 0 && gg_run pythia_2_8b + #test $ret -eq 0 && gg_run open_llama_7b_v2 fi test $ret -eq 0 && gg_run ctest_with_model_debug test $ret -eq 0 && gg_run ctest_with_model_release diff --git a/llama.cpp b/llama.cpp index 5ff186a57..1f9e10eed 100644 --- a/llama.cpp +++ b/llama.cpp @@ -6718,7 +6718,7 @@ static struct ggml_tensor * llm_build_kqv( cur = ggml_flash_attn_ext(ctx, q, k, v, kq_mask, kq_scale, hparams.f_max_alibi_bias); - if (model.arch == LLM_ARCH_PHI2 || model.arch == LLM_ARCH_PHI3) { + if (model.arch == LLM_ARCH_PHI2 || model.arch == LLM_ARCH_PHI3 || model.arch == LLM_ARCH_GPTNEOX) { ggml_flash_attn_ext_set_prec(cur, GGML_PREC_F32); } @@ -6727,7 +6727,7 @@ static struct ggml_tensor * llm_build_kqv( struct ggml_tensor * kq = ggml_mul_mat(ctx, k, q); cb(kq, "kq", il); - if (model.arch == LLM_ARCH_PHI2 || model.arch == LLM_ARCH_PHI3) { + if (model.arch == LLM_ARCH_PHI2 || model.arch == LLM_ARCH_PHI3 || model.arch == LLM_ARCH_GPTNEOX) { // for this arch, we need to perform the KQ multiplication with F32 precision, otherwise we get NaNs // ref: https://github.com/ggerganov/llama.cpp/pull/4490#issuecomment-1859055847 ggml_mul_mat_set_prec(kq, GGML_PREC_F32); From 3015851c5ac7334fb544a23a70a284c117b87044 Mon Sep 17 00:00:00 2001 From: Daniel Bevenius Date: Thu, 23 May 2024 14:29:26 +0200 Subject: [PATCH 3/4] llama : add getters for n_threads/n_threads_batch (#7464) * llama : add getters for n_threads/n_threads_batch This commit adds two new functions to the llama API. The functions can be used to get the number of threads used for generating a single token and the number of threads used for prompt and batch processing (multiple tokens). The motivation for this is that we want to be able to get the number of threads that the a context is using. The main use case is for a testing/verification that the number of threads is set correctly. Signed-off-by: Daniel Bevenius * squash! llama : add getters for n_threads/n_threads_batch Rename the getters to llama_n_threads and llama_n_threads_batch. Signed-off-by: Daniel Bevenius --------- Signed-off-by: Daniel Bevenius --- llama.cpp | 8 ++++++++ llama.h | 6 ++++++ 2 files changed, 14 insertions(+) diff --git a/llama.cpp b/llama.cpp index 1f9e10eed..e540c1b39 100644 --- a/llama.cpp +++ b/llama.cpp @@ -17410,6 +17410,14 @@ void llama_set_n_threads(struct llama_context * ctx, uint32_t n_threads, uint32_ ctx->cparams.n_threads_batch = n_threads_batch; } +uint32_t llama_n_threads(struct llama_context * ctx) { + return ctx->cparams.n_threads; +} + +uint32_t llama_n_threads_batch(struct llama_context * ctx) { + return ctx->cparams.n_threads_batch; +} + void llama_set_abort_callback(struct llama_context * ctx, bool (*abort_callback)(void * data), void * abort_callback_data) { ctx->abort_callback = abort_callback; ctx->abort_callback_data = abort_callback_data; diff --git a/llama.h b/llama.h index b7bf2afcb..16cece5db 100644 --- a/llama.h +++ b/llama.h @@ -759,6 +759,12 @@ extern "C" { // n_threads_batch is the number of threads used for prompt and batch processing (multiple tokens) LLAMA_API void llama_set_n_threads(struct llama_context * ctx, uint32_t n_threads, uint32_t n_threads_batch); + // Get the number of threads used for generation of a single token. + LLAMA_API uint32_t llama_n_threads(struct llama_context * ctx); + + // Get the number of threads used for prompt and batch processing (multiple token). + LLAMA_API uint32_t llama_n_threads_batch(struct llama_context * ctx); + // Set whether to use causal attention or not // If set to true, the model will only attend to the past tokens LLAMA_API void llama_set_causal_attn(struct llama_context * ctx, bool causal_attn); From 8b94e799dfa482adf63419df4905dc79b37e179f Mon Sep 17 00:00:00 2001 From: Raj Hammeer Singh Hada Date: Thu, 23 May 2024 18:00:13 +0530 Subject: [PATCH 4/4] readme : add Bunny in supported models [no ci] (#7469) --- README.md | 1 + 1 file changed, 1 insertion(+) diff --git a/README.md b/README.md index ccd12e211..461259afb 100644 --- a/README.md +++ b/README.md @@ -141,6 +141,7 @@ Typically finetunes of the base models below are supported as well. - [x] [Yi-VL](https://huggingface.co/models?search=Yi-VL) - [x] [Mini CPM](https://huggingface.co/models?search=MiniCPM) - [x] [Moondream](https://huggingface.co/vikhyatk/moondream2) +- [x] [Bunny](https://github.com/BAAI-DCAI/Bunny) **HTTP server**