---
license: other
model_name: Dolphin Llama 13B
base_model: ehartford/dolphin-llama-13b
inference: false
model_creator: Eric Hartford
model_type: llama
prompt_template: 'SYSTEM: {system_message}
USER: {prompt}
ASSISTANT:
'
quantized_by: TheBloke
task_categories:
- text-generation
---
# Dolphin Llama 13B - GPTQ
- Model creator: [Eric Hartford](https://huggingface.co/ehartford)
- Original model: [Dolphin Llama 13B](https://huggingface.co/ehartford/dolphin-llama-13b)
## Description
This repo contains GPTQ model files for [Eric Hartford's Dolphin Llama 13B](https://huggingface.co/ehartford/dolphin-llama-13b).
Multiple GPTQ parameter permutations are provided; see Provided Files below for details of the options provided, their parameters, and the software used to create them.
## Repositories available
* [AWQ model(s) for GPU inference.](https://huggingface.co/TheBloke/Dolphin-Llama-13B-AWQ)
* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGUF models for CPU+GPU inference](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GGUF)
* [Eric Hartford's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/ehartford/dolphin-llama-13b)
## Prompt template: Orca-Vicuna
```
SYSTEM: {system_message}
USER: {prompt}
ASSISTANT:
```
## Provided files and GPTQ parameters
Multiple quantisation parameters are provided, to allow you to choose the best one for your hardware and requirements.
Each separate quant is in a different branch. See below for instructions on fetching from different branches.
All recent GPTQ files are made with AutoGPTQ, and all files in non-main branches are made with AutoGPTQ. Files in the `main` branch which were uploaded before August 2023 were made with GPTQ-for-LLaMa.
Explanation of GPTQ parameters
- Bits: The bit size of the quantised model.
- GS: GPTQ group size. Higher numbers use less VRAM, but have lower quantisation accuracy. "None" is the lowest possible value.
- Act Order: True or False. Also known as `desc_act`. True results in better quantisation accuracy. Some GPTQ clients have had issues with models that use Act Order plus Group Size, but this is generally resolved now.
- Damp %: A GPTQ parameter that affects how samples are processed for quantisation. 0.01 is default, but 0.1 results in slightly better accuracy.
- GPTQ dataset: The dataset used for quantisation. Using a dataset more appropriate to the model's training can improve quantisation accuracy. Note that the GPTQ dataset is not the same as the dataset used to train the model - please refer to the original model repo for details of the training dataset(s).
- Sequence Length: The length of the dataset sequences used for quantisation. Ideally this is the same as the model sequence length. For some very long sequence models (16+K), a lower sequence length may have to be used. Note that a lower sequence length does not limit the sequence length of the quantised model. It only impacts the quantisation accuracy on longer inference sequences.
- ExLlama Compatibility: Whether this file can be loaded with ExLlama, which currently only supports Llama models in 4-bit.
| Branch | Bits | GS | Act Order | Damp % | GPTQ Dataset | Seq Len | Size | ExLlama | Desc |
| ------ | ---- | -- | --------- | ------ | ------------ | ------- | ---- | ------- | ---- |
| [main](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/main) | 4 | 128 | No | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 7.26 GB | Yes | 4-bit, without Act Order and group size 128g. |
| [gptq-4bit-32g-actorder_True](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-4bit-32g-actorder_True) | 4 | 32 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 8.00 GB | Yes | 4-bit, with Act Order and group size 32g. Gives highest possible inference quality, with maximum VRAM usage. |
| [gptq-4bit-64g-actorder_True](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-4bit-64g-actorder_True) | 4 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 7.51 GB | Yes | 4-bit, with Act Order and group size 64g. Uses less VRAM than 32g, but with slightly lower accuracy. |
| [gptq-4bit-128g-actorder_True](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-4bit-128g-actorder_True) | 4 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 7.26 GB | Yes | 4-bit, with Act Order and group size 128g. Uses even less VRAM than 64g, but with slightly lower accuracy. |
| [gptq-8bit--1g-actorder_True](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-8bit--1g-actorder_True) | 8 | None | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 13.36 GB | No | 8-bit, with Act Order. No group size, to lower VRAM requirements. |
| [gptq-8bit-128g-actorder_False](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-8bit-128g-actorder_False) | 8 | 128 | No | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 13.65 GB | No | 8-bit, with group size 128g for higher inference quality and without Act Order to improve AutoGPTQ speed. |
| [gptq-8bit-128g-actorder_True](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-8bit-128g-actorder_True) | 8 | 128 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 13.65 GB | No | 8-bit, with group size 128g for higher inference quality and with Act Order for even higher accuracy. |
| [gptq-8bit-64g-actorder_True](https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ/tree/gptq-8bit-64g-actorder_True) | 8 | 64 | Yes | 0.1 | [wikitext](https://huggingface.co/datasets/wikitext/viewer/wikitext-2-v1/test) | 2048 | 13.95 GB | No | 8-bit, with group size 64g and Act Order for even higher inference quality. Poor AutoGPTQ CUDA speed. |
## How to download from branches
- In text-generation-webui, you can add `:branch` to the end of the download name, eg `TheBloke/Dolphin-Llama-13B-GPTQ:main`
- With Git, you can clone a branch with:
```
git clone --single-branch --branch main https://huggingface.co/TheBloke/Dolphin-Llama-13B-GPTQ
```
- In Python Transformers code, the branch is the `revision` parameter; see below.
## How to easily download and use this model in [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
Please make sure you're using the latest version of [text-generation-webui](https://github.com/oobabooga/text-generation-webui).
It is strongly recommended to use the text-generation-webui one-click-installers unless you're sure you know how to make a manual install.
1. Click the **Model tab**.
2. Under **Download custom model or LoRA**, enter `TheBloke/Dolphin-Llama-13B-GPTQ`.
- To download from a specific branch, enter for example `TheBloke/Dolphin-Llama-13B-GPTQ:main`
- see Provided Files above for the list of branches for each option.
3. Click **Download**.
4. The model will start downloading. Once it's finished it will say "Done".
5. In the top left, click the refresh icon next to **Model**.
6. In the **Model** dropdown, choose the model you just downloaded: `Dolphin-Llama-13B-GPTQ`
7. The model will automatically load, and is now ready for use!
8. If you want any custom settings, set them and then click **Save settings for this model** followed by **Reload the Model** in the top right.
* Note that you do not need to and should not set manual GPTQ parameters any more. These are set automatically from the file `quantize_config.json`.
9. Once you're ready, click the **Text Generation tab** and enter a prompt to get started!
## How to use this GPTQ model from Python code
### Install the necessary packages
Requires: Transformers 4.32.0 or later, Optimum 1.12.0 or later, and AutoGPTQ 0.4.2 or later.
```shell
pip3 install transformers>=4.32.0 optimum>=1.12.0
pip3 install auto-gptq --extra-index-url https://huggingface.github.io/autogptq-index/whl/cu118/ # Use cu117 if on CUDA 11.7
```
If you have problems installing AutoGPTQ using the pre-built wheels, install it from source instead:
```shell
pip3 uninstall -y auto-gptq
git clone https://github.com/PanQiWei/AutoGPTQ
cd AutoGPTQ
pip3 install .
```
### For CodeLlama models only: you must use Transformers 4.33.0 or later.
If 4.33.0 is not yet released when you read this, you will need to install Transformers from source:
```shell
pip3 uninstall -y transformers
pip3 install git+https://github.com/huggingface/transformers.git
```
### You can then use the following code
```python
from transformers import AutoModelForCausalLM, AutoTokenizer, pipeline
model_name_or_path = "TheBloke/Dolphin-Llama-13B-GPTQ"
# To use a different branch, change revision
# For example: revision="main"
model = AutoModelForCausalLM.from_pretrained(model_name_or_path,
device_map="auto",
trust_remote_code=False,
revision="main")
tokenizer = AutoTokenizer.from_pretrained(model_name_or_path, use_fast=True)
prompt = "Tell me about AI"
prompt_template=f'''SYSTEM: {system_message}
USER: {prompt}
ASSISTANT:
'''
print("\n\n*** Generate:")
input_ids = tokenizer(prompt_template, return_tensors='pt').input_ids.cuda()
output = model.generate(inputs=input_ids, temperature=0.7, do_sample=True, top_p=0.95, top_k=40, max_new_tokens=512)
print(tokenizer.decode(output[0]))
# Inference can also be done using transformers' pipeline
print("*** Pipeline:")
pipe = pipeline(
"text-generation",
model=model,
tokenizer=tokenizer,
max_new_tokens=512,
do_sample=True,
temperature=0.7,
top_p=0.95,
top_k=40,
repetition_penalty=1.1
)
print(pipe(prompt_template)[0]['generated_text'])
```
## Compatibility
The files provided are tested to work with AutoGPTQ, both via Transformers and using AutoGPTQ directly. They should also work with [Occ4m's GPTQ-for-LLaMa fork](https://github.com/0cc4m/KoboldAI).
[ExLlama](https://github.com/turboderp/exllama) is compatible with Llama models in 4-bit. Please see the Provided Files table above for per-file compatibility.
[Huggingface Text Generation Inference (TGI)](https://github.com/huggingface/text-generation-inference) is compatible with all GPTQ models.
## Discord
For further support, and discussions on these models and AI in general, join us at:
[TheBloke AI's Discord server](https://discord.gg/theblokeai)
## Thanks, and how to contribute
Thanks to the [chirper.ai](https://chirper.ai) team!
Thanks to Clay from [gpus.llm-utils.org](llm-utils)!
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**: 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.
# Original model card: Eric Hartford's Dolphin Llama 13B
Dolphin 🐬
https://erichartford.com/dolphin
This model is based on llama1, so it is for non-commercial use only. Future versions will be trained on llama2 and other open models that are suitable for commercial use.
This model is uncensored. I have filtered the dataset to remove alignment and bias. This makes the model compliant to any requests. You are advised to implement your own alignment layer before exposing the model as a service. It will be highly compliant to any requests, even unethical ones. Please read my blog post about uncensored models. https://erichartford.com/uncensored-models
You are responsible for any content you create using this model. Enjoy responsibly.
## Dataset
This dataset is an open source implementation of [Microsoft's Orca](https://www.microsoft.com/en-us/research/publication/orca-progressive-learning-from-complex-explanation-traces-of-gpt-4/)
After uncensoring, deduping, and cleaning, our dataset consists of:
- 842,610 instructions of FLANv2 augmented with GPT-4 completions
- 2,625,353 instructions of FLANv2 augmented with GPT-3.5 completions
We followed the submix and system prompt distribution outlined in the Orca paper. With a few exceptions. We included all 75k of CoT in the FLAN-1m dataset rather than sampling that. Also, we found that many items were duplicated, so we removed duplicates.
Then we filtered out instances of alignment, refusal, avoidance, and bias, in order to produce an uncensored model upon which can be layered your personalized alignment LoRA.
We also filtered out duplicates and cleaned the data.
## Training
We trained with the flan5m (gpt3.5 completions) dataset in its entirety for 3 epochs at a learning rate of 2e-5 before we stopped training to avoid overfit.
We trained with the flan1m (gpt4 completions) dataset in its entirety for 2.5 epochs at a learning rate of 1e-5 before we stopped training to avoid overfit.
It took about 600 hours to train on 8x H100s
We used a prompt format similar to Vicuna, but we added the SYSTEM: field.
Prompt format:
```
SYSTEM: {system}
USER: {prompt}
ASSISTANT:
```
Example:
```
SYSTEM: you are an expert marine biologist.
USER: Please list 10 ways that dolphins are superior to orcas.
ASSISTANT:
```
## Evaluation
Evaluation will be coming soon.
## Team
The core Dolphin Team includes:
- Eric "Faldore" Hartford
- Pankaj Mathur
- Rob "Rohan" O'Callahan
- Tom "TheBloke" Jobbins
## Gratitude
- Thank you to Microsoft for authoring the Orca paper and inspiring this work.
- Special thanks to WingLian, NanoBit, Teknium for helpful advice
- Special thanks to EdenCoder and chirper.ai for mentorship and financial sponsorship.
- Special thanks to Kilkonie for his very valued mentorship.
- Thank you to Catto.
- Thank you to Nicolai Schleifer, financial sponsor.
- Thank you to Eric Fleming, financial sponsor.
- Thank you to all the other people in the Open Source AI community who have taught me and helped me along the way.
- And biggest thanks and love to my wife Hope and my babies, who have endured two months of my absenteeism while working long days and nights on this project.
## Example Output
```
SYSTEM: you are an expert marine biologist.
USER: Please list 10 ways that dolphins are superior to orcas.
ASSISTANT:
```
1. Social structure: Dolphins live in complex social groups called pods, which consist of extended family units and can include hundreds of individuals. Orcas, on the other hand, typically form smaller, more temporary groupings known as clans or matrilineal kinship groups.
2. Communication: Dolphins have a highly developed system of vocalizations, body language, and even telepathy to communicate with each other for various purposes such as coordinating cooperative hunting strategies, sharing information about food sources, and maintaining social bonds. While orcas also exhibit some communication abilities, their methods are generally less sophisticated compared to those of dolphins.
3. Intelligence: Dolphins are widely regarded as one of the most intelligent animal species, showcasing advanced problem-solving skills, self-awareness, tool use (such as using sea sponges as tools), and learning capabilities. Some dolphin species have been observed demonstrating altruism and understanding of human emotions. In contrast, while orcas are also smart, they do not display the same level of cognitive complexity as dolphins.
4. Playfulness: Dolphins are well-known for their playful behavior, often engaging in acrobatic displays, chasing each other's tails, and playing with objects like balls or seaweed. This playful nature is thought to be essential for their social bonding and mental stimulation. Although orcas may occasionally interact playfully, it is not as prevalent or diverse as among dolphins.
5. Agility and speed: Dolphins are incredibly agile swimmers, capable of reaching speeds up to 30 miles per hour (48 km/h) when pursuing prey. They can change direction rapidly and maneuver through tight spaces with ease. While orcas are powerful predators and fast swimmers themselves, they cannot match the sheer acceleration and agility of dolphins.
6. Feeding habits: Both dolphins and orcas are carnivorous, feeding primarily on fish and cephalopods. However, dolphins employ a wider variety of hunting techniques, including cooperative hunting strategies involving herding, encircling, and driving schools of fish into shallow waters where they become disoriented and easier to catch. Orcas mainly rely on ambushing their prey from below and using their powerful jaws to strike quickly.
7. Reproduction: Female dolphins usually give birth to a single calf after a gestation period of around 12 months. The mother plays a crucial role in raising and nurturing her offspring until they become independent. On the other hand, female orcas tend to breed every three to five years, giving birth to multiple offspring at once, and relying heavily on their mothers for protection and teaching.
8. Vocal repertoire: Dolphins possess a wide range of sounds, including clicks, whistles, and body sounds, which allow them to convey different messages and establish unique identities within their pods. Orcas, too, produce vocalizations but lack the diversity found in dolphin calls.
9. Conservation status: Many dolphin species face threats due to habitat loss, pollution, overfishing, and climate change. As a result, several populations are considered vulnerable or endangered by the International Union for Conservation of Nature (IUCN). In comparison, orca conservation status varies depending on the region; however, no species is currently listed as critically endangered or facing extinction.
10. Adaptability: Dolphins have adapted to various aquatic environments, ranging from warm coastal waters to colder open oceans. Their versatility allows them to thrive in both tropical and temperate climates. Orcas, conversely, are adapted to specific habitats such as cooler coastal waters and are mostly found in the Northern Hemisphere.