Downloads · 30 days
3.6K
9% of all-time downloads
TheBloke/Llama2-70B-OASST-SFT-v10-GGUF
Llama2-70B-OASST-SFT-v10-GGUF is a text generation model from TheBloke. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as llama2.
<div style="width: auto; margin-left: auto; margin-right: auto" <img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;" </div <div style="displa…
Downloads · 30 days
3.6K
9% of all-time downloads
All-time downloads
40.5K
Public
Repo size
959 GB
Likes
10
Public
Click a slice to open those files.
.gguf392 GB · 75%
From the Hugging Face model README
This repo contains GGUF format model files for OpenAssistant's Llama2 70B SFT v10.
<!-- description end --> <!-- README_GGUF.md-about-gguf start -->GGUF is a new format introduced by the llama.cpp team on August 21st 2023. It is a replacement for GGML, which is no longer supported by llama.cpp. GGUF offers numerous advantages over GGML, such as better tokenisation, and support for special tokens. It is also supports metadata, and is designed to be extensible.
Here is an incomplate list of clients and libraries that are known to support GGUF:
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
<!-- prompt-template end -->
<!-- compatibility_gguf start -->
These quantised GGUFv2 files are compatible with llama.cpp from August 27th onwards, as of commit d0cee0d36d5be95a0d9088b674dbb27354107221
They are also compatible with many third party UIs and libraries - please see the list at the top of this README.
The new methods available are:
Refer to the Provided Files table below to see what files use which methods, and how.
</details> <!-- compatibility_gguf end --> <!-- README_GGUF.md-provided-files start -->| Name | Quant method | Bits | Size | Max RAM required | Use case |
|---|---|---|---|---|---|
| llama2-70b-oasst-sft-v10.Q2_K.gguf | Q2_K | 2 | 29.28 GB | 31.78 GB | smallest, significant quality loss - not recommended for most purposes |
| llama2-70b-oasst-sft-v10.Q3_K_S.gguf | Q3_K_S | 3 | 29.92 GB | 32.42 GB | very small, high quality loss |
| llama2-70b-oasst-sft-v10.Q3_K_M.gguf | Q3_K_M | 3 | 33.19 GB | 35.69 GB | very small, high quality loss |
| llama2-70b-oasst-sft-v10.Q3_K_L.gguf | Q3_K_L | 3 | 36.15 GB | 38.65 GB | small, substantial quality loss |
| llama2-70b-oasst-sft-v10.Q4_0.gguf | Q4_0 | 4 | 38.87 GB | 41.37 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| llama2-70b-oasst-sft-v10.Q4_K_S.gguf | Q4_K_S | 4 | 39.08 GB | 41.58 GB | small, greater quality loss |
| llama2-70b-oasst-sft-v10.Q4_K_M.gguf | Q4_K_M | 4 | 41.42 GB | 43.92 GB | medium, balanced quality - recommended |
| llama2-70b-oasst-sft-v10.Q5_0.gguf | Q5_0 | 5 | 47.46 GB | 49.96 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| llama2-70b-oasst-sft-v10.Q5_K_S.gguf | Q5_K_S | 5 | 47.46 GB | 49.96 GB | large, low quality loss - recommended |
| llama2-70b-oasst-sft-v10.Q5_K_M.gguf | Q5_K_M | 5 | 48.76 GB | 51.26 GB | large, very low quality loss - recommended |
| llama2-70b-oasst-sft-v10.Q6_K.gguf | Q6_K | 6 | 56.59 GB | 59.09 GB | very large, extremely low quality loss |
| llama2-70b-oasst-sft-v10.Q8_0.gguf | Q8_0 | 8 | 73.29 GB | 75.79 GB | very large, extremely low quality loss - not recommended |
Note: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
Note: HF does not support uploading files larger than 50GB. Therefore I have uploaded the Q6_K and Q8_0 files as split files.
<details> <summary>Click for instructions regarding Q6_K and Q8_0 files</summary>Please download:
llama2-70b-oasst-sft-v10.Q6_K.gguf-split-allama2-70b-oasst-sft-v10.Q6_K.gguf-split-bPlease download:
llama2-70b-oasst-sft-v10.Q8_0.gguf-split-allama2-70b-oasst-sft-v10.Q8_0.gguf-split-bTo join the files, do the following:
Linux and macOS:
cat llama2-70b-oasst-sft-v10.Q6_K.gguf-split-* > llama2-70b-oasst-sft-v10.Q6_K.gguf && rm llama2-70b-oasst-sft-v10.Q6_K.gguf-split-*
cat llama2-70b-oasst-sft-v10.Q8_0.gguf-split-* > llama2-70b-oasst-sft-v10.Q8_0.gguf && rm llama2-70b-oasst-sft-v10.Q8_0.gguf-split-*
Windows command line:
COPY /B llama2-70b-oasst-sft-v10.Q6_K.gguf-split-a + llama2-70b-oasst-sft-v10.Q6_K.gguf-split-b llama2-70b-oasst-sft-v10.Q6_K.gguf
del llama2-70b-oasst-sft-v10.Q6_K.gguf-split-a llama2-70b-oasst-sft-v10.Q6_K.gguf-split-b
COPY /B llama2-70b-oasst-sft-v10.Q8_0.gguf-split-a + llama2-70b-oasst-sft-v10.Q8_0.gguf-split-b llama2-70b-oasst-sft-v10.Q8_0.gguf
del llama2-70b-oasst-sft-v10.Q8_0.gguf-split-a llama2-70b-oasst-sft-v10.Q8_0.gguf-split-b
</details>
<!-- README_GGUF.md-provided-files end -->
<!-- README_GGUF.md-how-to-download start -->
Note for manual downloaders: You almost never want to clone the entire repo! Multiple different quantisation formats are provided, and most users only want to pick and download a single file.
The following clients/libraries will automatically download models for you, providing a list of available models to choose from:
text-generation-webuiUnder Download Model, you can enter the model repo: TheBloke/Llama2-70B-OASST-SFT-v10-GGUF and below it, a specific filename to download, such as: llama2-70b-oasst-sft-v10.q4_K_M.gguf.
Then click Download.
I recommend using the huggingface-hub Python library:
pip3 install huggingface-hub>=0.17.1
Then you can download any individual model file to the current directory, at high speed, with a command like this:
huggingface-cli download TheBloke/Llama2-70B-OASST-SFT-v10-GGUF llama2-70b-oasst-sft-v10.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
<details>
<summary>More advanced huggingface-cli download usage</summary>
You can also download multiple files at once with a pattern:
huggingface-cli download TheBloke/Llama2-70B-OASST-SFT-v10-GGUF --local-dir . --local-dir-use-symlinks False --include='*Q4_K*gguf'
For more documentation on downloading with huggingface-cli, please see: HF -> Hub Python Library -> Download files -> Download from the CLI.
To accelerate downloads on fast connections (1Gbit/s or higher), install hf_transfer:
pip3 install hf_transfer
And set environment variable HF_HUB_ENABLE_HF_TRANSFER to 1:
HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download TheBloke/Llama2-70B-OASST-SFT-v10-GGUF llama2-70b-oasst-sft-v10.q4_K_M.gguf --local-dir . --local-dir-use-symlinks False
Windows CLI users: Use set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 before running the download command.
llama.cpp commandMake sure you are using llama.cpp from commit d0cee0d36d5be95a0d9088b674dbb27354107221 or later.
./main -ngl 32 -m llama2-70b-oasst-sft-v10.q4_K_M.gguf --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "<|im_start|>system\n{system_message}<|im_end|>\n<|im_start|>user\n{prompt}<|im_end|>\n<|im_start|>assistant"
Change -ngl 32 to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.
Change -c 4096 to the desired sequence length. For extended sequence models - eg 8K, 16K, 32K - the necessary RoPE scaling parameters are read from the GGUF file and set by llama.cpp automatically.
If you want to have a chat-style conversation, replace the -p <PROMPT> argument with -i -ins
For other parameters and how to use them, please refer to the llama.cpp documentation
text-generation-webuiFurther instructions here: text-generation-webui/docs/llama.cpp.md.
You can use GGUF models from Python using the llama-cpp-python or ctransformers libraries.
# Base ctransformers with no GPU acceleration
pip install ctransformers>=0.2.24
# Or with CUDA GPU acceleration
pip install ctransformers[cuda]>=0.2.24
# Or with ROCm GPU acceleration
CT_HIPBLAS=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
# Or with Metal GPU acceleration for macOS systems
CT_METAL=1 pip install ctransformers>=0.2.24 --no-binary ctransformers
from ctransformers import AutoModelForCausalLM
# Set gpu_layers to the number of layers to offload to GPU. Set to 0 if no GPU acceleration is available on your system.
llm = AutoModelForCausalLM.from_pretrained("TheBloke/Llama2-70B-OASST-SFT-v10-GGUF", model_file="llama2-70b-oasst-sft-v10.q4_K_M.gguf", model_type="llama", gpu_layers=50)
print(llm("AI is going to"))
Here's guides on using llama-cpp-python or ctransformers with LangChain:
<!-- README_GGUF.md-how-to-run end --> <!-- footer start --> <!-- 200823 -->For further support, and discussions on these models and AI in general, join us at:
Thanks to the chirper.ai team!
Thanks to Clay from gpus.llm-utils.org!
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
Special thanks to: Aemon Algiz.
Patreon special mentions: Alicia Loh, Stephen Murray, K, Ajan Kanaga, RoA, Magnesian, Deo Leter, Olakabola, Eugene Pentland, zynix, Deep Realms, Raymond Fosdick, Elijah Stavena, Iucharbius, Erik Bjäreholt, Luis Javier Navarrete Lozano, Nicholas, theTransient, John Detwiler, alfie_i, knownsqashed, Mano Prime, Willem Michiel, Enrico Ros, LangChain4j, OG, Michael Dempsey, Pierre Kircher, Pedro Madruga, James Bentley, Thomas Belote, Luke @flexchar, Leonard Tan, Johann-Peter Hartmann, Illia Dulskyi, Fen Risland, Chadd, S_X, Jeff Scroggin, Ken Nordquist, Sean Connelly, Artur Olbinski, Swaroop Kallakuri, Jack West, Ai Maven, David Ziegler, Russ Johnson, transmissions 11, John Villwock, Alps Aficionado, Clay Pascal, Viktor Bowallius, Subspace Studios, Rainer Wilmers, Trenton Dambrowitz, vamX, Michael Levine, 준교 김, Brandon Frisco, Kalila, Trailburnt, Randy H, Talal Aujan, Nathan Dryer, Vadim, 阿明, ReadyPlayerEmma, Tiffany J. Kim, George Stoitzev, Spencer Kim, Jerry Meng, Gabriel Tamborski, Cory Kujawski, Jeffrey Morgan, Spiking Neurons AB, Edmond Seymore, Alexandros Triantafyllidis, Lone Striker, Cap'n Zoog, Nikolai Manek, danny, ya boyyy, Derek Yates, usrbinkat, Mandus, TL, Nathan LeClaire, subjectnull, Imad Khwaja, webtim, Raven Klaugh, Asp the Wyvern, Gabriel Puliatti, Caitlyn Gatomon, Joseph William Delisle, Jonathan Leane, Luke Pendergrass, SuperWojo, Sebastain Graf, Will Dee, Fred von Graf, Andrey, Dan Guido, Daniel P. Andersen, Nitin Borwankar, Elle, Vitor Caleffi, biorpg, jjj, NimbleBox.ai, Pieter, Matthew Berman, terasurfer, Michael Davis, Alex, Stanislav Ovsiannikov
Thank you to all my generous patrons and donaters!
And thank you again to a16z for their generous grant.
<!-- footer end --> <!-- original-model-card start -->This model is an Open-Assistant fine-tuning of Meta's Llama2 70B LLM. It was fine-tuned in two stages, first on a mix of synthetic instrunctions and coding tasks and then in a "polishing" stage on the best human demonstrations collected at open-assistant.io up to July 23, 2023 (see Configuration Details below).
Due to public demand (see survey) we changed the prompt-template for this model from custom prompter/assistant tokens to OpenAI's chatml standard prompt format. We hope that this leads to greater compatibility with chat inference/frontend applications.
Prompt dialogue template:
"""
<|im_start|>system
{system_message}<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant
"""
The model input can contain multiple conversation turns between user and assistant, e.g.
<|im_start|>user
{prompt 1}<|im_end|>
<|im_start|>assistant
{reply 1}<|im_end|>
<|im_start|>user
{prompt 2}<|im_end|>
<|im_start|>assistant
(...)
The model was partly trained with orca system messages. For inference we recommend to use the official Llama2 system message:
<|im_start|>system
You are a helpful, respectful and honest assistant. Always answer as helpfully as possible, while being safe. Your answers should not include any harmful, unethical, racist, sexist, toxic, dangerous, or illegal content. Please ensure that your responses are socially unbiased and positive in nature.
If a question does not make any sense, or is not factually coherent, explain why instead of answering something not correct. If you don't know the answer to a question, please don't share false information.
<|im_end|>
We want to especially thank everyone who contributed in the crowed-sourced Open-Assistant dataset creation on https://open-assistant.io/ - without you this project would not have been possible.
Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, the potential outputs of llama2-70b-oasst-sft-v10 cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of llama2-70b-oasst-sft-v10, developers should perform safety testing and tuning tailored to their specific applications of the model.
Please see Meta's Responsible Use Guide.
During evaluation we noticed that this 70B model produced extremely poor outputs when loaded it was loaded in 16 bit precision sharded in TGI. In contrast the model could be evaluated without problem using vLLM. The model also worked decently well when loaded with TGI on a single GPPU nf4 quantized via TimDettmers/bitsandbytes. Will will get it touch with the TGI authors to find out why sharded 16-bit inference doesn't work as expected.
The "pretokenizer" utility used to tokenize the datamix is part of the Open-Assistant github repository and can be found here: model/pretokenizer.
Entries of the dataset with assistant replies shorter than 25 tokens were excluded from training.
oasst_pre10_min25:
datasets:
- megacode2:
fraction: 0.5
val_split: 0.01
max_val_set: 1000
- orca-chat:
val_split: 0.01
max_val_set: 1000
- dolly15k_multilingual:
val_split: 0.05
max_val_set: 300
- oa_leet10k:
val_split: 0.05
max_val_set: 250
output_dir: "output/oasst_pre10_min25"
filename_prefix: "oasst_pre10"
min_assistant_tokens: 25
Stage 1 dataset statistics:
# Stats for output/oasst_pre10_min25_llama2
## Stats for 'Subset of InstructionDataset (megacode2)' (466364 samples (50.0%))
-----------------
Accepted: 398223/466364 (85.4%)
Accepted tokens: 167676873
Skipped: 68141 (14.6%)
Min tokens per sample: 36
Max tokens per sample: 11810
Avg tokens per sample: 421.063
-----------------
## Stats for 'Subset of OrcaChat (orca-chat)' (325616 samples (100.0%))
-----------------
Accepted: 325616/325616 (100.0%)
Accepted tokens: 178307574
Skipped: 0 (0.0%)
Min tokens per sample: 105
Max tokens per sample: 10408
Avg tokens per sample: 547.601
-----------------
## Stats for 'Subset of Dolly15kMultilingual' (57020 samples (100.0%))
-----------------
Accepted: 47494/57020 (83.3%)
Accepted tokens: 13883177
Skipped: 9526 (16.7%)
Min tokens per sample: 34
Max tokens per sample: 9172
Avg tokens per sample: 292.314
-----------------
## Stats for 'Subset of InstructionDataset (oa_leet10k)' (22236 samples (100.0%))
-----------------
Accepted: 22236/22236 (100.0%)
Accepted tokens: 15905296
Skipped: 0 (0.0%)
Min tokens per sample: 168
Max tokens per sample: 10588
Avg tokens per sample: 715.295
-----------------
## Stats for 'total' (871236 samples (100.0%))
-----------------
Accepted: 793569/871236 (91.1%)
Accepted tokens: 375772920
Skipped: 77667 (8.9%)
Min tokens per sample: 34
Max tokens per sample: 11810
Avg tokens per sample: 473.523
-----------------
oasst_top1:
datasets:
- oasst_export:
lang: "bg,ca,cs,da,de,en,es,fr,hr,hu,it,nl,pl,pt,ro,ru,sl,sr,sv,uk"
input_file_path: 2023-07-23_oasst_ready.tar.gz
top_k: 1
val_split: 0.05
output_dir: "output/oasst_top1_2023-07-23"
filename_prefix: "oasst_top1"
Stage 2 dataset statistics:
# Stats for output/oasst_top1_2023-07-23_llama2
## Stats for 'ListDataset' (11441 samples (100.0%))
-----------------
Accepted: 11441/11441 (100.0%)
Accepted tokens: 5315368
Skipped: 0 (0.0%)
Min tokens per sample: 20
Max tokens per sample: 5407
Avg tokens per sample: 464.58945896337735
-----------------
## Stats for 'total' (11441 samples (100.0%))
-----------------
Accepted: 11441/11441 (100.0%)
Accepted tokens: 5315368
Skipped: 0 (0.0%)
Min tokens per sample: 20
Max tokens per sample: 5407
Avg tokens per sample: 464.58945896337735
-----------------
--tensor_model_parallel_size 8
--pipeline_model_parallel_size 4
--load ./checkpoints/llama2-70b-tp8-pp4
--save ./checkpoints/llama2-70b-tp8-pp4-oasst_pre10
--tensorboard_dir ./checkpoints/llama2-70b-tp8-pp4-oasst_pre10/logging
--data_path ./data/oasst_pre10_min25_llama2/oasst_sft10-train
--model_name llama2
--tokenizer_type SentencePieceTokenizer
--bf16
--global_batch_size 64
--micro_batch_size 2
--vocab_file=./llama2/Llama-2-7b/tokenizer.model
--use_rms_norm
--glu_activation swiglu
--no_tie_embed_logits
--vocab_extra_ids_list "\"<|im_start|>,<|im_end|>\""
--layernorm_epsilon 1e-5
--use_flash_attn
--no_bias_gelu_fusion
--seq_length 4096
--max_position_embeddings 4096
--log_interval 1
--save_interval 500
--eval_interval 50
--eval_iters 10
--hidden_dropout 0.0
--position_embedding_type rotary
--no_bias_dropout_fusion
--use_checkpoint_args
--train_iters 12000
--attention_dropout 0.0
--adam_beta1 0.9
--adam_beta2 0.95
--adam_eps 1e-12
--lr_decay_style cosine
--lr_warmup_iters 100
--lr 1e-5
--min_lr 1e-6
--weight_decay 0.000001
--sequence_parallel
--recompute_granularity selective
--log_timers_to_tensorboard
--rope_scaling_factor 1.0
--wandb_logger
--tensor_model_parallel_size 8
--pipeline_model_parallel_size 4
--load ./checkpoints/llama2-70b-tp8-pp4-oasst_pre10
--save ./checkpoints/llama2-70b-tp8-pp4-oasst_sft10
--tensorboard_dir ./checkpoints/llama2-70b-tp8-pp4-oasst_sft10/logging
--data_path ./data/oasst_top1_2023-07-23_llama2/oasst_top1-train
--model_name llama2
--tokenizer_type SentencePieceTokenizer
--bf16
--global_batch_size 64
--micro_batch_size 2
--vocab_file=./llama2/Llama-2-7b/tokenizer.model
--use_rms_norm
--glu_activation swiglu
--no_tie_embed_logits
--vocab_extra_ids_list "\"<|im_start|>,<|im_end|>\""
--layernorm_epsilon 1e-5
--use_flash_attn
--no_bias_gelu_fusion
--seq_length 4096
--max_position_embeddings 4096
--log_interval 1
--save_interval 346
--eval_interval 50
--eval_iters 10
--hidden_dropout 0.25
--lima_dropout
--position_embedding_type rotary
--no_bias_dropout_fusion
--use_checkpoint_args
--train_iters 519
--attention_dropout 0.0
--adam_beta1 0.9
--adam_beta2 0.95
--adam_eps 1e-12
--lr_decay_style cosine
--lr_warmup_iters 100
--lr 1e-5
--min_lr 1e-6
--weight_decay 0.000001
--sequence_parallel
--recompute_granularity selective
--log_timers_to_tensorboard
--rope_scaling_factor 1.0
--finetune
--wandb_logger
<!-- original-model-card end -->