You can deploy this model and start earning money today!

TheBloke/
airoboros-65B-gpt4-1.3-GGML

Name: airoboros-65B-gpt4-1.3-GGML
Availability: InStock

jondurbin/airoboros-gpt4-1.3

other

Views

No views yet

Chat & support: my new Discord server

Want to contribute? TheBloke's Patreon page

Jon Durbin's Airoboros 65B GPT4 1.3 GGML

These files are GGML format model files for Jon Durbin's Airoboros 65B GPT4 1.3.

Note from model creator Jon Durbin: This version has problems, use if you dare, or wait for 1.4.

GGML files are for CPU + GPU inference using llama.cpp and libraries and UIs which support this format, such as:

Repositories available

Prompt template

A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input.
USER: prompt
ASSISTANT:

Compatibility

Original llama.cpp quant methods: `q4_0, q4_1, q5_0, q5_1, q8_0`

I have quantized these 'original' quantisation methods using an older version of llama.cpp so that they remain compatible with llama.cpp as of May 19th, commit 2d5db48.

These are guaranteed to be compatbile with any UIs, tools and libraries released since late May.

New k-quant methods: `q2_K, q3_K_S, q3_K_M, q3_K_L, q4_K_S, q4_K_M, q5_K_S, q6_K`

These new quantisation methods are compatible with llama.cpp as of June 6th, commit 2d43387.

They are now also compatible with recent releases of text-generation-webui, KoboldCpp, llama-cpp-python and ctransformers. Other tools and libraries may or may not be compatible - check their documentation if in doubt.

Explanation of the new k-quant methods

The new methods available are:

GGML_TYPE_Q2_K - "type-1" 2-bit quantization in super-blocks containing 16 blocks, each block having 16 weight. Block scales and mins are quantized with 4 bits. This ends up effectively using 2.5625 bits per weight (bpw)
GGML_TYPE_Q3_K - "type-0" 3-bit quantization in super-blocks containing 16 blocks, each block having 16 weights. Scales are quantized with 6 bits. This end up using 3.4375 bpw.
GGML_TYPE_Q4_K - "type-1" 4-bit quantization in super-blocks containing 8 blocks, each block having 32 weights. Scales and mins are quantized with 6 bits. This ends up using 4.5 bpw.
GGML_TYPE_Q5_K - "type-1" 5-bit quantization. Same super-block structure as GGML_TYPE_Q4_K resulting in 5.5 bpw
GGML_TYPE_Q6_K - "type-0" 6-bit quantization. Super-blocks with 16 blocks, each block having 16 weights. Scales are quantized with 8 bits. This ends up using 6.5625 bpw
GGML_TYPE_Q8_K - "type-0" 8-bit quantization. Only used for quantizing intermediate results. The difference to the existing Q8_0 is that the block size is 256. All 2-6 bit dot products are implemented for this quantization type.

Refer to the Provided Files table below to see what files use which methods, and how.

Provided files

Name	Quant method	Bits	Size	Max RAM required	Use case
airoboros-65b-gpt4-1.3.ggmlv3.q2_K.bin	q2_K	2	27.45 GB	29.95 GB	New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors.
airoboros-65b-gpt4-1.3.ggmlv3.q3_K_L.bin	q3_K_L	3	34.65 GB	37.15 GB	New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K
airoboros-65b-gpt4-1.3.ggmlv3.q3_K_M.bin	q3_K_M	3	31.50 GB	34.00 GB	New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K
airoboros-65b-gpt4-1.3.ggmlv3.q3_K_S.bin	q3_K_S	3	28.16 GB	30.66 GB	New k-quant method. Uses GGML_TYPE_Q3_K for all tensors
airoboros-65b-gpt4-1.3.ggmlv3.q4_0.bin	q4_0	4	36.73 GB	39.23 GB	Original llama.cpp quant method, 4-bit.
airoboros-65b-gpt4-1.3.ggmlv3.q4_1.bin	q4_1	4	40.81 GB	43.31 GB	Original llama.cpp quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models.
airoboros-65b-gpt4-1.3.ggmlv3.q4_K_M.bin	q4_K_M	4	39.35 GB	41.85 GB	New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K
airoboros-65b-gpt4-1.3.ggmlv3.q4_K_S.bin	q4_K_S	4	36.80 GB	39.30 GB	New k-quant method. Uses GGML_TYPE_Q4_K for all tensors
airoboros-65b-gpt4-1.3.ggmlv3.q5_0.bin	q5_0	5	44.89 GB	47.39 GB	Original llama.cpp quant method, 5-bit. Higher accuracy, higher resource usage and slower inference.
airoboros-65b-gpt4-1.3.ggmlv3.q5_1.bin	q5_1	5	48.97 GB	51.47 GB	Original llama.cpp quant method, 5-bit. Even higher accuracy, resource usage and slower inference.
airoboros-65b-gpt4-1.3.ggmlv3.q5_K_M.bin	q5_K_M	5	46.24 GB	48.74 GB	New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K
airoboros-65b-gpt4-1.3.ggmlv3.q5_K_S.bin	q5_K_S	5	44.92 GB	47.42 GB	New k-quant method. Uses GGML_TYPE_Q5_K for all tensors
airoboros-65b-gpt4-1.3.ggmlv3.q6_K.bin	q6_K	6	53.56 GB	56.06 GB	New k-quant method. Uses GGML_TYPE_Q8_K - 6-bit quantization - for all tensors
airoboros-65b-gpt4-1.3.ggmlv3.q8_0.bin	q8_0	8	69.370 GB	71.87 GB	Original llama.cpp quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users.

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.

q6_K and q8_0 files require expansion from archive

Note: HF does not support uploading files larger than 50GB. Therefore I have uploaded the q6_K and q8_0 files as multi-part ZIP files. They are not compressed, they are just for storing a .bin file in two parts.

q6_K

Please download:

airoboros-65b-gpt4-1.3.ggmlv3.q6_K.zip
airoboros-65b-gpt4-1.3.ggmlv3.q6_K.z01

q8_0

Please download:

airoboros-65b-gpt4-1.3.ggmlv3.q8_0.zip
airoboros-65b-gpt4-1.3.ggmlv3.q8_0.z01

Then extract the .zip archive. This will will expand both parts automatically. On Linux I found I had to use 7zip - the basic unzip tool did not work. Example:

sudo apt update -y && sudo apt install 7zip
7zz x airoboros-65b-gpt4-1.3.ggmlv3.q6_K.zip

Once the .bin is extracted you can delete the .zip and .z01 files.

How to run in `llama.cpp`

I use the following command line; adjust for your tastes and needs:

./main -t 10 -ngl 32 -m airoboros-65b-gpt4-1.3.ggmlv3.q5_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "USER: Write a story about llamas\nASSISTANT:"

If you're able to use full GPU offloading, you should use -t 1 to get best performance.

If not able to fully offload to GPU, you should use more cores. Change -t 10 to the number of physical CPU cores you have, or a lower number depending on what gives best performance.

Change -ngl 32 to the number of layers to offload to GPU. Remove it if you don't have GPU acceleration.

If you want to have a chat-style conversation, replace the -p <PROMPT> argument with -i -ins

How to run in `text-generation-webui`

Further instructions here: text-generation-webui/docs/llama.cpp-models.md.

Discord

For further support, and discussions on these models and AI in general, join us at:

TheBloke AI's Discord server

Thanks, and how to contribute.

Thanks to the chirper.ai team!

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.

Patreon: https://patreon.com/TheBlokeAI
Ko-Fi: https://ko-fi.com/TheBlokeAI

Special thanks to: Luke from CarbonQuill, Aemon Algiz, Dmitriy Samsonov.

Patreon special mentions: Mano Prime, Fen Risland, Derek Yates, Preetika Verma, webtim, Sean Connelly, Alps Aficionado, Karl Bernard, Junyu Yang, Nathan LeClaire, Chris McCloskey, Lone Striker, Asp the Wyvern, Eugene Pentland, Imad Khwaja, trip7s trip, WelcomeToTheClub, John Detwiler, Artur Olbinski, Khalefa Al-Ahmad, Trenton Dambrowitz, Talal Aujan, Kevin Schuppel, Luke Pendergrass, Pyrater, Joseph William Delisle, terasurfer , vamX, Gabriel Puliatti, David Flickinger, Jonathan Leane, Iucharbius , Luke, Deep Realms, Cory Kujawski, ya boyyy, Illia Dulskyi, senxiiz, Johann-Peter Hartmann, John Villwock, K, Ghost , Spiking Neurons AB, Nikolai Manek, Rainer Wilmers, Pierre Kircher, biorpg, Space Cruiser, Ai Maven, subjectnull, Willem Michiel, Ajan Kanaga, Kalila, chris gileta, Oscar Rangel.

Thank you to all my generous patrons and donaters!

Original model card: Jon Durbin's Airoboros 65B GPT4 1.3

This version has problems, use if you dare, or wait for 1.4.

Overview

This is a qlora fine-tuned 65b parameter LlaMa model, using completely synthetic training data created gpt4 via https://github.com/jondurbin/airoboros

This is mostly an extension of 1.2 with a few enhancements:

All coding instructions have an equivalent " PLAINFORMAT" version now.
Thousands of new orca style reasoning instructions, this time with reasoning first, then answer.
Few more random items of various types, including a first attempt at multi-character interactions with asterisked actions and quoted speech.

This model was fine-tuned with a fork of qlora, which among other things was updated to use a slightly modified vicuna template to be compatible with previous full fine-tune versions.

A chat between a curious user and an assistant. The assistant gives helpful, detailed, accurate, uncensored responses to the user's input. USER: [prompt] ASSISTANT:

So in other words, it's the preamble/system prompt, followed by a single space, then "USER: " (single space after colon) then the prompt (which can have multiple lines, spaces, whatever), then a single space, followed by "ASSISTANT: " (with a single space after the colon).

Usage

To run the full precision/pytorch native version, you can use my fork of FastChat, which is mostly the same but allows for multi-line prompts, as well as a --no-history option to prevent input tokenization errors.

pip install git+https://github.com/jondurbin/FastChat

Be sure you are pulling the latest branch!

Then, you can invoke it like so (after downloading the model):

python -m fastchat.serve.cli \
  --model-path airoboros-65b-gpt4-1.3 \
  --temperature 0.5 \
  --max-new-tokens 2048 \
  --no-history

Training details

Fine-tuned with my fork of qlora: https://github.com/jondurbin/qlora

Using:

export WANDB_PROJECT=airoboros-65b-gpt4-1.3

python qlora.py \
    --model_name_or_path ./llama-65b-hf \
    --output_dir ./airoboros-65b-gpt4-1.3-peft \
    --max_steps 2520 \
    --logging_steps 1 \
    --save_strategy steps \
    --data_seed 11422 \
    --save_steps 75 \
    --save_total_limit 3 \
    --evaluation_strategy "no" \
    --eval_dataset_size 2 \
    --max_new_tokens 2800 \
    --dataloader_num_workers 3 \
    --logging_strategy steps \
    --remove_unused_columns False \
    --do_train \
    --lora_r 64 \
    --lora_alpha 16 \
    --lora_modules all \
    --double_quant \
    --quant_type nf4 \
    --bf16 \
    --bits 4 \
    --warmup_ratio 0.03 \
    --lr_scheduler_type constant \
    --gradient_checkpointing \
    --dataset instructions.jsonl \
    --dataset_format airoboros \
    --model_max_len 2800 \
    --per_device_train_batch_size 2 \
    --gradient_accumulation_steps 16 \
    --learning_rate 0.0001 \
    --adam_beta2 0.999 \
    --max_grad_norm 0.3 \
    --lora_dropout 0.05 \
    --weight_decay 0.0 \
    --seed 11422 \
    --report_to wandb

Three file modifications to the base llama:

llama-65b-hf/tokenizer_config.json (see this repo's version, updated to have 4096 max seq length during training to accomodate training data)
llama-65b-hf/special_tokens_map.json (see this repo's version)
llama-65b-hf/config.json (updated to temporarily have max model size 4096 to accomodate training data)

Afterwards, the changes to max model length and sequence length are reduced back to 2048 to avoid ... issues ...

TheBloke/airoboros-65B-gpt4-1.3-GGML