Chat & support: jartine's Discord server

jartine's LLM work is generously supported by a grant from mozilla

---

Nous Hermes Llama 2 13B - llamafile

• Model creator: NousResearch
• Original model: Nous Hermes Llama 2 13B

Description

This repo contains llamafile format model files for Nous Research's Nous Hermes Llama 2 13B.

WARNING: This README may contain inaccuracies. It was generated automatically by forking TheBloke/Nous-Hermes-Llama2-GGUF and piping the README through sed. Errors should be reported to jartine, and do not reflect TheBloke. You can support his work on Patreon.

About llamafile

llamafile is a new format introduced by Mozilla Ocho on Nov 20th 2023. It uses Cosmopolitan Libc to turn LLM weights into runnable llama.cpp binaries that run on the stock installs of six OSes for both ARM64 and AMD64. llamafile 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 llamafile:

• llama.cpp. The source project for llamafile. Offers a CLI and a server option.
• text-generation-webui, the most widely used web UI, with many features and powerful extensions. Supports GPU acceleration.
• KoboldCpp, a fully featured web UI, with GPU accel across all platforms and GPU architectures. Especially good for story telling.
• LM Studio, an easy-to-use and powerful local GUI for Windows and macOS (Silicon), with GPU acceleration.
• LoLLMS Web UI, a great web UI with many interesting and unique features, including a full model library for easy model selection.
• Faraday.dev, an attractive and easy to use character-based chat GUI for Windows and macOS (both Silicon and Intel), with GPU acceleration.
• ctransformers, a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
• llama-cpp-python, a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.
• candle, a Rust ML framework with a focus on performance, including GPU support, and ease of use.

Repositories available

• AWQ model(s) for GPU inference.
• GPTQ models for GPU inference, with multiple quantisation parameter options.
• 2, 3, 4, 5, 6 and 8-bit llamafile models for CPU+GPU inference
• NousResearch's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions

Prompt template: Alpaca

Below is an instruction that describes a task. Write a response that appropriately completes the request.

### Instruction:
{prompt}

### Response:

Licensing

The creator of the source model has listed its license as ['mit'], and this quantization has therefore used that same license.

As this model is based on Llama 2, it is also subject to the Meta Llama 2 license terms, and the license files for that are additionally included. It should therefore be considered as being claimed to be licensed under both licenses. I contacted Hugging Face for clarification on dual licensing but they do not yet have an official position. Should this change, or should Meta provide any feedback on this situation, I will update this section accordingly.

In the meantime, any questions regarding licensing, and in particular how these two licenses might interact, should be directed to the original model repository: Nous Research's Nous Hermes Llama 2 13B.

Compatibility

These quantised llamafilev2 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.

Explanation of quantisation methods

Click to see details

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

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

Provided files

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.

How to download llamafile files

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:

• LM Studio
• LoLLMS Web UI
• Faraday.dev

In text-generation-webui

Under Download Model, you can enter the model repo: jartine/Nous-Hermes-Llama2-llamafile and below it, a specific filename to download, such as: nous-hermes-llama2-13b.q4_K_M.llamafile.

Then click Download.

On the command line, including multiple files at once

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 jartine/Nous-Hermes-Llama2-llamafile nous-hermes-llama2-13b.q4_K_M.llamafile --local-dir . --local-dir-use-symlinks False

More advanced huggingface-cli download usage

You can also download multiple files at once with a pattern:

huggingface-cli download jartine/Nous-Hermes-Llama2-llamafile --local-dir . --local-dir-use-symlinks False --include='*Q4_K*llamafile'

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 jartine/Nous-Hermes-Llama2-llamafile nous-hermes-llama2-13b.q4_K_M.llamafile --local-dir . --local-dir-use-symlinks False

Windows CLI users: Use set HUGGINGFACE_HUB_ENABLE_HF_TRANSFER=1 before running the download command.

Example llama.cpp command

Make sure you are using llama.cpp from commit d0cee0d36d5be95a0d9088b674dbb27354107221 or later.

./main -ngl 32 -m nous-hermes-llama2-13b.q4_K_M.llamafile --color -c 4096 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request.\n\n### Instruction:\n{prompt}\n\n### Response:"

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 llamafile 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

How to run in text-generation-webui

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

How to run from Python code

You can use llamafile models from Python using the llama-cpp-python or ctransformers libraries.

How to load this model from Python using ctransformers

First install the package

# 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

Simple example code to load one of these llamafile models

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("jartine/Nous-Hermes-Llama2-llamafile", model_file="nous-hermes-llama2-13b.q4_K_M.llamafile", model_type="llama", gpu_layers=50)

print(llm("AI is going to"))

How to use with LangChain

Here's guides on using llama-cpp-python or ctransformers with LangChain:

• LangChain + llama-cpp-python
• LangChain + ctransformers

Discord

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

jartine AI's Discord server

Thanks, and how to contribute

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.

And thank you again to mozilla for their generous grant.

Original model card: Nous Research's Nous Hermes Llama 2 13B

Model Card: Nous-Hermes-Llama2-13b

Compute provided by our project sponsor Redmond AI, thank you! Follow RedmondAI on Twitter @RedmondAI.

Model Description

Nous-Hermes-Llama2-13b is a state-of-the-art language model fine-tuned on over 300,000 instructions. This model was fine-tuned by Nous Research, with Teknium and Emozilla leading the fine tuning process and dataset curation, Redmond AI sponsoring the compute, and several other contributors.

This Hermes model uses the exact same dataset as Hermes on Llama-1. This is to ensure consistency between the old Hermes and new, for anyone who wanted to keep Hermes as similar to the old one, just more capable.

This model stands out for its long responses, lower hallucination rate, and absence of OpenAI censorship mechanisms. The fine-tuning process was performed with a 4096 sequence length on an 8x a100 80GB DGX machine.

Example Outputs:

Model Training

The model was trained almost entirely on synthetic GPT-4 outputs. Curating high quality GPT-4 datasets enables incredibly high quality in knowledge, task completion, and style.

This includes data from diverse sources such as GPTeacher, the general, roleplay v1&2, code instruct datasets, Nous Instruct & PDACTL (unpublished), and several others, detailed further below

Collaborators

The model fine-tuning and the datasets were a collaboration of efforts and resources between Teknium, Karan4D, Emozilla, Huemin Art, and Redmond AI.

Special mention goes to @winglian for assisting in some of the training issues.

Huge shoutout and acknowledgement is deserved for all the dataset creators who generously share their datasets openly.

Among the contributors of datasets:

• GPTeacher was made available by Teknium
• Wizard LM by nlpxucan
• Nous Research Instruct Dataset was provided by Karan4D and HueminArt.
• GPT4-LLM and Unnatural Instructions were provided by Microsoft
• Airoboros dataset by jondurbin
• Camel-AI's domain expert datasets are from Camel-AI
• CodeAlpaca dataset by Sahil 2801.

If anyone was left out, please open a thread in the community tab.

Prompt Format

The model follows the Alpaca prompt format:

### Instruction:
<prompt>

### Response:
<leave a newline blank for model to respond>

or

### Instruction:
<prompt>

### Input:
<additional context>

### Response:
<leave a newline blank for model to respond>

Benchmark Results

AGI-Eval

|             Task             |Version| Metric |Value |   |Stderr|
|agieval_aqua_rat              |      0|acc     |0.2362|±  |0.0267|
|                              |       |acc_norm|0.2480|±  |0.0272|
|agieval_logiqa_en             |      0|acc     |0.3425|±  |0.0186|
|                              |       |acc_norm|0.3472|±  |0.0187|
|agieval_lsat_ar               |      0|acc     |0.2522|±  |0.0287|
|                              |       |acc_norm|0.2087|±  |0.0269|
|agieval_lsat_lr               |      0|acc     |0.3510|±  |0.0212|
|                              |       |acc_norm|0.3627|±  |0.0213|
|agieval_lsat_rc               |      0|acc     |0.4647|±  |0.0305|
|                              |       |acc_norm|0.4424|±  |0.0303|
|agieval_sat_en                |      0|acc     |0.6602|±  |0.0331|
|                              |       |acc_norm|0.6165|±  |0.0340|
|agieval_sat_en_without_passage|      0|acc     |0.4320|±  |0.0346|
|                              |       |acc_norm|0.4272|±  |0.0345|
|agieval_sat_math              |      0|acc     |0.2909|±  |0.0307|
|                              |       |acc_norm|0.2727|±  |0.0301|

GPT-4All Benchmark Set

|    Task     |Version| Metric |Value |   |Stderr|
|arc_challenge|      0|acc     |0.5102|±  |0.0146|
|             |       |acc_norm|0.5213|±  |0.0146|
|arc_easy     |      0|acc     |0.7959|±  |0.0083|
|             |       |acc_norm|0.7567|±  |0.0088|
|boolq        |      1|acc     |0.8394|±  |0.0064|
|hellaswag    |      0|acc     |0.6164|±  |0.0049|
|             |       |acc_norm|0.8009|±  |0.0040|
|openbookqa   |      0|acc     |0.3580|±  |0.0215|
|             |       |acc_norm|0.4620|±  |0.0223|
|piqa         |      0|acc     |0.7992|±  |0.0093|
|             |       |acc_norm|0.8069|±  |0.0092|
|winogrande   |      0|acc     |0.7127|±  |0.0127|

BigBench Reasoning Test

|                      Task                      |Version|       Metric        |Value |   |Stderr|

|bigbench_causal_judgement                       |      0|multiple_choice_grade|0.5526|±  |0.0362|
|bigbench_date_understanding                     |      0|multiple_choice_grade|0.7344|±  |0.0230|
|bigbench_disambiguation_qa                      |      0|multiple_choice_grade|0.2636|±  |0.0275|
|bigbench_geometric_shapes                       |      0|multiple_choice_grade|0.0195|±  |0.0073|
|                                                |       |exact_str_match      |0.0000|±  |0.0000|
|bigbench_logical_deduction_five_objects         |      0|multiple_choice_grade|0.2760|±  |0.0200|
|bigbench_logical_deduction_seven_objects        |      0|multiple_choice_grade|0.2100|±  |0.0154|
|bigbench_logical_deduction_three_objects        |      0|multiple_choice_grade|0.4400|±  |0.0287|
|bigbench_movie_recommendation                   |      0|multiple_choice_grade|0.2440|±  |0.0192|
|bigbench_navigate                               |      0|multiple_choice_grade|0.4950|±  |0.0158|
|bigbench_reasoning_about_colored_objects        |      0|multiple_choice_grade|0.5570|±  |0.0111|
|bigbench_ruin_names                             |      0|multiple_choice_grade|0.3728|±  |0.0229|
|bigbench_salient_translation_error_detection    |      0|multiple_choice_grade|0.1854|±  |0.0123|
|bigbench_snarks                                 |      0|multiple_choice_grade|0.6298|±  |0.0360|
|bigbench_sports_understanding                   |      0|multiple_choice_grade|0.6156|±  |0.0155|
|bigbench_temporal_sequences                     |      0|multiple_choice_grade|0.3140|±  |0.0147|
|bigbench_tracking_shuffled_objects_five_objects |      0|multiple_choice_grade|0.2032|±  |0.0114|
|bigbench_tracking_shuffled_objects_seven_objects|      0|multiple_choice_grade|0.1406|±  |0.0083|
|bigbench_tracking_shuffled_objects_three_objects|      0|multiple_choice_grade|0.4400|±  |0.0287|

These are the highest benchmarks Hermes has seen on every metric, achieving the following average scores:

• GPT4All benchmark average is now 70.0 - from 68.8 in Hermes-Llama1
• 0.3657 on BigBench, up from 0.328 on hermes-llama1
• 0.372 on AGIEval, up from 0.354 on Hermes-llama1

These benchmarks currently have us at #1 on ARC-c, ARC-e, Hellaswag, and OpenBookQA, and 2nd place on Winogrande, comparing to GPT4all's benchmarking list, supplanting Hermes 1 for the new top position.

Resources for Applied Use Cases:

Check out LM Studio for a nice chatgpt style interface here: https://lmstudio.ai/ For an example of a back and forth chatbot using huggingface transformers and discord, check out: https://github.com/teknium1/alpaca-discord For an example of a roleplaying discord chatbot, check out this: https://github.com/teknium1/alpaca-roleplay-discordbot

Future Plans

We plan to continue to iterate on both more high quality data, and new data filtering techniques to eliminate lower quality data going forward.

Model Usage

The model is available for download on Hugging Face. It is suitable for a wide range of language tasks, from generating creative text to understanding and following complex instructions.