add files
Browse files- .gitattributes +1 -0
- .gitignore +8 -0
- Makefile +55 -0
- README.md +12 -0
- SHA256SUMS +15 -0
- open-llama-13b-f16.gguf +3 -0
- open-llama-13b-q2_K.gguf +3 -0
- open-llama-13b-q3_K_L.gguf +3 -0
- open-llama-13b-q3_K_M.gguf +3 -0
- open-llama-13b-q3_K_S.gguf +3 -0
- open-llama-13b-q4_0.gguf +3 -0
- open-llama-13b-q4_1.gguf +3 -0
- open-llama-13b-q4_K_M.gguf +3 -0
- open-llama-13b-q4_K_S.gguf +3 -0
- open-llama-13b-q5_0.gguf +3 -0
- open-llama-13b-q5_1.gguf +3 -0
- open-llama-13b-q5_K_M.gguf +3 -0
- open-llama-13b-q5_K_S.gguf +3 -0
- open-llama-13b-q6_K.gguf +3 -0
- open-llama-13b-q8_0.gguf +3 -0
.gitattributes
CHANGED
|
@@ -4,6 +4,7 @@
|
|
| 4 |
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
*.ftz filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 7 |
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 8 |
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 9 |
*.joblib filter=lfs diff=lfs merge=lfs -text
|
|
|
|
| 4 |
*.bz2 filter=lfs diff=lfs merge=lfs -text
|
| 5 |
*.ckpt filter=lfs diff=lfs merge=lfs -text
|
| 6 |
*.ftz filter=lfs diff=lfs merge=lfs -text
|
| 7 |
+
*.gguf filter=lfs diff=lfs merge=lfs -text
|
| 8 |
*.gz filter=lfs diff=lfs merge=lfs -text
|
| 9 |
*.h5 filter=lfs diff=lfs merge=lfs -text
|
| 10 |
*.joblib filter=lfs diff=lfs merge=lfs -text
|
.gitignore
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
llama.cpp*/
|
| 2 |
+
venv/
|
| 3 |
+
pytorch_model*.bin
|
| 4 |
+
*.sha
|
| 5 |
+
*.tar.gz
|
| 6 |
+
tokenizer.model
|
| 7 |
+
config.json
|
| 8 |
+
tokenizer_config.json
|
Makefile
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
MODEL_NAME= open-llama-13b
|
| 2 |
+
PYTHON?= python
|
| 3 |
+
LLAMA_BUILD= 1132
|
| 4 |
+
LLAMA_TAR= b$(LLAMA_BUILD).tar.gz
|
| 5 |
+
LLAMA_DIR= llama.cpp-b$(LLAMA_BUILD)
|
| 6 |
+
LLAMA_FLAGS=
|
| 7 |
+
HF_REPO= openlm-research/open_llama_13b
|
| 8 |
+
HF_REF= main
|
| 9 |
+
HF_FILES= pytorch_model-00001-of-00003.bin \
|
| 10 |
+
pytorch_model-00002-of-00003.bin \
|
| 11 |
+
pytorch_model-00003-of-00003.bin \
|
| 12 |
+
tokenizer.model \
|
| 13 |
+
config.json \
|
| 14 |
+
tokenizer_config.json
|
| 15 |
+
$(HF_FILES): SITE= https://huggingface.co/$(HF_REPO)/resolve/$(HF_REF)
|
| 16 |
+
$(LLAMA_TAR): SITE= https://github.com/ggerganov/llama.cpp/archive/refs/tags
|
| 17 |
+
|
| 18 |
+
QUANTS= f16 q4_0 q4_1 q5_0 q5_1 q8_0 \
|
| 19 |
+
q2_K \
|
| 20 |
+
q3_K_S q3_K_M q3_K_L \
|
| 21 |
+
q4_K_S q4_K_M \
|
| 22 |
+
q5_K_S q5_K_M \
|
| 23 |
+
q6_K
|
| 24 |
+
|
| 25 |
+
FILES= $(HF_FILES) $(LLAMA_TAR)
|
| 26 |
+
MODEL_FILES= $(foreach q,$(QUANTS),$(MODEL_NAME)-$(q).gguf)
|
| 27 |
+
|
| 28 |
+
.PHONY: all
|
| 29 |
+
all: $(MODEL_FILES) SHA256SUMS
|
| 30 |
+
|
| 31 |
+
$(FILES):
|
| 32 |
+
curl -L -o $@ --url $(SITE)/$@
|
| 33 |
+
|
| 34 |
+
$(LLAMA_DIR): | $(LLAMA_TAR)
|
| 35 |
+
tar -xf $(LLAMA_TAR)
|
| 36 |
+
|
| 37 |
+
$(LLAMA_DIR)/quantize: | $(LLAMA_DIR)
|
| 38 |
+
$(MAKE) -C $(LLAMA_DIR) $(LLAMA_FLAGS) quantize
|
| 39 |
+
|
| 40 |
+
venv:
|
| 41 |
+
$(PYTHON) -m venv venv
|
| 42 |
+
venv/bin/pip install -e $(LLAMA_DIR)/gguf-py
|
| 43 |
+
venv/bin/pip install -r $(LLAMA_DIR)/requirements.txt
|
| 44 |
+
|
| 45 |
+
$(MODEL_NAME)-f16.gguf: $(HF_FILES) | $(LLAMA_DIR) venv
|
| 46 |
+
venv/bin/python $(LLAMA_DIR)/convert.py --outtype f16 --outfile $@ .
|
| 47 |
+
|
| 48 |
+
$(MODEL_NAME)-q%.gguf: $(MODEL_NAME)-f16.gguf | $(LLAMA_DIR)/quantize
|
| 49 |
+
$(LLAMA_DIR)/quantize $< $@ q$*
|
| 50 |
+
|
| 51 |
+
%.sha: %
|
| 52 |
+
sha256sum $< > $@
|
| 53 |
+
|
| 54 |
+
SHA256SUMS: $(addsuffix .sha,$(MODEL_FILES))
|
| 55 |
+
cat $^ > $@
|
README.md
CHANGED
|
@@ -1,3 +1,15 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
# gguf versions of OpenLLaMa 13B
|
| 6 |
+
|
| 7 |
+
- Version: 1000B tokens final release
|
| 8 |
+
- Project: [OpenLLaMA: An Open Reproduction of LLaMA](https://github.com/openlm-research/open_llama)
|
| 9 |
+
- Model: [openlm-research/open_llama_13b](https://huggingface.co/openlm-research/open_llama_13b)
|
| 10 |
+
- [llama.cpp](https://github.com/ggerganov/llama.cpp): build 1012 (6381d4e) or later
|
| 11 |
+
- [ggml version](https://huggingface.co/SlyEcho/open_llama_13b_ggml)
|
| 12 |
+
|
| 13 |
+
## Perplexity on wiki.test.406
|
| 14 |
+
|
| 15 |
+
Coming soon...
|
SHA256SUMS
ADDED
|
@@ -0,0 +1,15 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
2fe1ca2677c8d44af840d87201e2648357a67e9ebbb40db42fc8d36c82b182ef open-llama-13b-f16.gguf
|
| 2 |
+
1bd67ba3ad2d34a776a8c2cca336106fbc2ec20428d15c3f7ae35a9d2cc996db open-llama-13b-q4_0.gguf
|
| 3 |
+
03dcf76d23f34dd0064bf87d0eb3c34cd9c371c4ab80629f5acca0bbbf438dac open-llama-13b-q4_1.gguf
|
| 4 |
+
0175ae7413f49f8aae62cd070b5456d45ac02bf9f436b99655e3a6df31088108 open-llama-13b-q5_0.gguf
|
| 5 |
+
12791fa6b9409f7b11dd960e9e29da665d3180c83a4d2749c77816dc3bf6ab5e open-llama-13b-q5_1.gguf
|
| 6 |
+
9afa5144b42c6f8686c2d2e4cddd4716ffce3db173d30c864b0e225c9fb79251 open-llama-13b-q8_0.gguf
|
| 7 |
+
9dc971555a9616fe453cdb47c233f16fc9a4f3645b4519b7603c49d55e1b11b1 open-llama-13b-q2_K.gguf
|
| 8 |
+
010c957b686d1fb8928dcc5af320d20f415faec35edd9ad5bc753c728088311f open-llama-13b-q3_K_S.gguf
|
| 9 |
+
ac3601bbdb2b08fd1d7a98d13f1ee6bcaa9e3c5d38814e520d1b863e086c6c3e open-llama-13b-q3_K_M.gguf
|
| 10 |
+
2d377a89d2f223174969b009738afa9343c85d44d8aad900ba3b78be41a2dcda open-llama-13b-q3_K_L.gguf
|
| 11 |
+
3a59166e77242001175217a00ec5fa75aa4babdb51ec95b0f74bd5843c2b8fe1 open-llama-13b-q4_K_S.gguf
|
| 12 |
+
cf40b5d26f362646f81f9c119dec16331568f4b377f7153d262b173639dd6d1c open-llama-13b-q4_K_M.gguf
|
| 13 |
+
70495f5db2fa068684f80bda421f5654306ea4c760fd7b86cc2f09059431b92d open-llama-13b-q5_K_S.gguf
|
| 14 |
+
8a07e7c7b6e8b2d8fc33332c4d4cc3d9454b4b2fdb1ff7d147d3a699e10c4e49 open-llama-13b-q5_K_M.gguf
|
| 15 |
+
e228650322158a84810924066ca4a8fdf88f8df1804d82a509aabcea881297b9 open-llama-13b-q6_K.gguf
|
open-llama-13b-f16.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2fe1ca2677c8d44af840d87201e2648357a67e9ebbb40db42fc8d36c82b182ef
|
| 3 |
+
size 26033337888
|
open-llama-13b-q2_K.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9dc971555a9616fe453cdb47c233f16fc9a4f3645b4519b7603c49d55e1b11b1
|
| 3 |
+
size 5429382752
|
open-llama-13b-q3_K_L.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2d377a89d2f223174969b009738afa9343c85d44d8aad900ba3b78be41a2dcda
|
| 3 |
+
size 6929593952
|
open-llama-13b-q3_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ac3601bbdb2b08fd1d7a98d13f1ee6bcaa9e3c5d38814e520d1b863e086c6c3e
|
| 3 |
+
size 6337803872
|
open-llama-13b-q3_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:010c957b686d1fb8928dcc5af320d20f415faec35edd9ad5bc753c728088311f
|
| 3 |
+
size 5659014752
|
open-llama-13b-q4_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1bd67ba3ad2d34a776a8c2cca336106fbc2ec20428d15c3f7ae35a9d2cc996db
|
| 3 |
+
size 7365869152
|
open-llama-13b-q4_1.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:03dcf76d23f34dd0064bf87d0eb3c34cd9c371c4ab80629f5acca0bbbf438dac
|
| 3 |
+
size 8169094752
|
open-llama-13b-q4_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cf40b5d26f362646f81f9c119dec16331568f4b377f7153d262b173639dd6d1c
|
| 3 |
+
size 7865990752
|
open-llama-13b-q4_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3a59166e77242001175217a00ec5fa75aa4babdb51ec95b0f74bd5843c2b8fe1
|
| 3 |
+
size 7414365792
|
open-llama-13b-q5_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0175ae7413f49f8aae62cd070b5456d45ac02bf9f436b99655e3a6df31088108
|
| 3 |
+
size 8972320352
|
open-llama-13b-q5_1.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:12791fa6b9409f7b11dd960e9e29da665d3180c83a4d2749c77816dc3bf6ab5e
|
| 3 |
+
size 9775545952
|
open-llama-13b-q5_K_M.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8a07e7c7b6e8b2d8fc33332c4d4cc3d9454b4b2fdb1ff7d147d3a699e10c4e49
|
| 3 |
+
size 9229958752
|
open-llama-13b-q5_K_S.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:70495f5db2fa068684f80bda421f5654306ea4c760fd7b86cc2f09059431b92d
|
| 3 |
+
size 8972320352
|
open-llama-13b-q6_K.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e228650322158a84810924066ca4a8fdf88f8df1804d82a509aabcea881297b9
|
| 3 |
+
size 10679174752
|
open-llama-13b-q8_0.gguf
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9afa5144b42c6f8686c2d2e4cddd4716ffce3db173d30c864b0e225c9fb79251
|
| 3 |
+
size 13831353952
|