File size: 27,963 Bytes

8e547b7
 
 
 
7ea0abc
 
 
 
8e547b7
 
 
 
 
7ea0abc
8e547b7

---
language:
- en
license: apache-2.0
datasets:
- totally-not-an-llm/everything-sharegptformat-morecleaned
model_name: Marx 3B
inference: false
model_creator: Bohan Du
model_link: https://huggingface.co/acrastt/marx-3b
model_type: llama
pipeline_tag: text-generation
quantized_by: TheBloke
base_model: acrastt/marx-3b
---

<!-- header start -->
<!-- 200823 -->
<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="display: flex; justify-content: space-between; width: 100%;">
    <div style="display: flex; flex-direction: column; align-items: flex-start;">
        <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://discord.gg/theblokeai">Chat & support: TheBloke's Discord server</a></p>
    </div>
    <div style="display: flex; flex-direction: column; align-items: flex-end;">
        <p style="margin-top: 0.5em; margin-bottom: 0em;"><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
    </div>
</div>
<div style="text-align:center; margin-top: 0em; margin-bottom: 0em"><p style="margin-top: 0.25em; margin-bottom: 0em;">TheBloke's LLM work is generously supported by a grant from <a href="https://a16z.com">andreessen horowitz (a16z)</a></p></div>
<hr style="margin-top: 1.0em; margin-bottom: 1.0em;">
<!-- header end -->

# Marx 3B - GGML
- Model creator: [Bohan Du](https://huggingface.co/acrastt)
- Original model: [Marx 3B](https://huggingface.co/acrastt/marx-3b)

## Description

This repo contains GGML format model files for [Bohan Du's Marx 3B](https://huggingface.co/acrastt/marx-3b).

GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most popular web UI. Supports NVidia CUDA GPU acceleration.
* [KoboldCpp](https://github.com/LostRuins/koboldcpp), a powerful GGML web UI with GPU acceleration on all platforms (CUDA and OpenCL). Especially good for story telling.
* [LM Studio](https://lmstudio.ai/), a fully featured local GUI with GPU acceleration on both Windows (NVidia and AMD), and macOS.
* [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with CUDA GPU acceleration via the c_transformers backend.
* [ctransformers](https://github.com/marella/ctransformers), a Python library with GPU accel, LangChain support, and OpenAI-compatible AI server.
* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with GPU accel, LangChain support, and OpenAI-compatible API server.

## Repositories available

* [GPTQ models for GPU inference, with multiple quantisation parameter options.](https://huggingface.co/TheBloke/Marx-3b-GPTQ)
* [2, 3, 4, 5, 6 and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/Marx-3b-GGML)
* [Bohan Du's original unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/acrastt/marx-3b)

## Prompt template: Human-Response

```
### HUMAN:
{prompt}

### RESPONSE:
```

<!-- compatibility_ggml start -->
## Compatibility

These quantised GGML files are compatible with llama.cpp as of June 6th, commit `2d43387`.

They should also be compatible with all UIs, libraries and utilities which use GGML.

## Explanation of the new k-quant methods
<details>
  <summary>Click to see details</summary>

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.
</details>
<!-- compatibility_ggml end -->

## Provided files

| Name | Quant method | Bits | Size | Max RAM required | Use case |
| ---- | ---- | ---- | ---- | ---- | ----- |
| [marx-3b.ggmlv3.q4_0.bin](https://huggingface.co/TheBloke/Marx-3b-GGML/blob/main/marx-3b.ggmlv3.q4_0.bin) | q4_0 | 4 | 1.93 GB| 4.43 GB | Original quant method, 4-bit. |
| [marx-3b.ggmlv3.q4_1.bin](https://huggingface.co/TheBloke/Marx-3b-GGML/blob/main/marx-3b.ggmlv3.q4_1.bin) | q4_1 | 4 | 2.14 GB| 4.64 GB | Original quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
| [marx-3b.ggmlv3.q5_0.bin](https://huggingface.co/TheBloke/Marx-3b-GGML/blob/main/marx-3b.ggmlv3.q5_0.bin) | q5_0 | 5 | 2.36 GB| 4.86 GB | Original quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
| [marx-3b.ggmlv3.q5_1.bin](https://huggingface.co/TheBloke/Marx-3b-GGML/blob/main/marx-3b.ggmlv3.q5_1.bin) | q5_1 | 5 | 2.57 GB| 5.07 GB | Original quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
| [marx-3b.ggmlv3.q8_0.bin](https://huggingface.co/TheBloke/Marx-3b-GGML/blob/main/marx-3b.ggmlv3.q8_0.bin) | q8_0 | 8 | 3.64 GB| 6.14 GB | Original 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.

## How to run in `llama.cpp`

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

```
./main -t 10 -ngl 32 -m marx-3b.ggmlv3.q4_K_M.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### Instruction: Write a story about llamas\n### Response:"
```
Change `-t 10` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`.

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

Change `-c 2048` to the desired sequence length for this model. For example, `-c 4096` for a Llama 2 model.  For models that use RoPE, add `--rope-freq-base 10000 --rope-freq-scale 0.5` for doubled context, or `--rope-freq-base 10000 --rope-freq-scale 0.25` for 4x context.

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](https://github.com/ggerganov/llama.cpp/blob/master/examples/main/README.md)

## How to run in `text-generation-webui`

Further instructions here: [text-generation-webui/docs/llama.cpp.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp.md).

<!-- footer start -->
<!-- 200823 -->
## 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!

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**: Sam, theTransient, Jonathan Leane, Steven Wood, webtim, Johann-Peter Hartmann, Geoffrey Montalvo, Gabriel Tamborski, Willem Michiel, John Villwock, Derek Yates, Mesiah Bishop, Eugene Pentland, Pieter, Chadd, Stephen Murray, Daniel P. Andersen, terasurfer, Brandon Frisco, Thomas Belote, Sid, Nathan LeClaire, Magnesian, Alps Aficionado, Stanislav Ovsiannikov, Alex, Joseph William Delisle, Nikolai Manek, Michael Davis, Junyu Yang, K, J, Spencer Kim, Stefan Sabev, Olusegun Samson, transmissions 11, Michael Levine, Cory Kujawski, Rainer Wilmers, zynix, Kalila, Luke @flexchar, Ajan Kanaga, Mandus, vamX, Ai Maven, Mano Prime, Matthew Berman, subjectnull, Vitor Caleffi, Clay Pascal, biorpg, alfie_i, 阿明, Jeffrey Morgan, ya boyyy, Raymond Fosdick, knownsqashed, Olakabola, Leonard Tan, ReadyPlayerEmma, Enrico Ros, Dave, Talal Aujan, Illia Dulskyi, Sean Connelly, senxiiz, Artur Olbinski, Elle, Raven Klaugh, Fen Risland, Deep Realms, Imad Khwaja, Fred von Graf, Will Dee, usrbinkat, SuperWojo, Alexandros Triantafyllidis, Swaroop Kallakuri, Dan Guido, John Detwiler, Pedro Madruga, Iucharbius, Viktor Bowallius, Asp the Wyvern, Edmond Seymore, Trenton Dambrowitz, Space Cruiser, Spiking Neurons AB, Pyrater, LangChain4j, Tony Hughes, Kacper Wikieł, Rishabh Srivastava, David Ziegler, Luke Pendergrass, Andrey, Gabriel Puliatti, Lone Striker, Sebastain Graf, Pierre Kircher, Randy H, NimbleBox.ai, Vadim, danny, Deo Leter


Thank you to all my generous patrons and donaters!

And thank you again to a16z for their generous grant.

<!-- footer end -->

# Original model card: Bohan Du's Marx 3B

This is [OpenLLaMA 3B V2](https://huggingface.co/openlm-research/open_llama_3b_v2) finetuned on [EverythingLM Data(ShareGPT format more cleaned)](https://huggingface.co/datasets/totally-not-an-llm/everything-sharegptformat-morecleaned) for 1 epochs.

Prompt template:
```
### HUMAN:
{prompt}

### RESPONSE:
<leave a newline for the model to answer>
```
q4_1 GGML quant [here](https://huggingface.co/NikolayKozloff/EverythingLM-3B).</br>
All GGML quants available [here](https://huggingface.co/asedmammad/Marx-3B-GGML).

Note: Don't expect this model to be good, I was just starting out to finetune. So don't roast me please!

Benchmarks:<table><tr><th>arc_challenge</th><td><table><tr><th>acc</th><td>0.38993174061433444</td></tr><tr><th>acc_stderr</th><td>0.014252959848892884</td></tr><tr><th>acc_norm</th><td>0.4308873720136519</td></tr><tr><th>acc_norm_stderr</th><td>0.014471133392642475</td></tr></table></td></tr><tr><th>hellaswag</th><td><table><tr><th>acc</th><td>0.5513841864170484</td></tr><tr><th>acc_stderr</th><td>0.004963362085275556</td></tr><tr><th>acc_norm</th><td>0.7257518422624976</td></tr><tr><th>acc_norm_stderr</th><td>0.00445222854104355</td></tr></table></td></tr><tr><th>hendrycksTest-abstract_algebra</th><td><table><tr><th>acc</th><td>0.23</td></tr><tr><th>acc_stderr</th><td>0.04229525846816506</td></tr><tr><th>acc_norm</th><td>0.23</td></tr><tr><th>acc_norm_stderr</th><td>0.04229525846816506</td></tr></table></td></tr><tr><th>hendrycksTest-anatomy</th><td><table><tr><th>acc</th><td>0.2962962962962963</td></tr><tr><th>acc_stderr</th><td>0.03944624162501116</td></tr><tr><th>acc_norm</th><td>0.2962962962962963</td></tr><tr><th>acc_norm_stderr</th><td>0.03944624162501116</td></tr></table></td></tr><tr><th>hendrycksTest-astronomy</th><td><table><tr><th>acc</th><td>0.32894736842105265</td></tr><tr><th>acc_stderr</th><td>0.03823428969926603</td></tr><tr><th>acc_norm</th><td>0.32894736842105265</td></tr><tr><th>acc_norm_stderr</th><td>0.03823428969926603</td></tr></table></td></tr><tr><th>hendrycksTest-business_ethics</th><td><table><tr><th>acc</th><td>0.3</td></tr><tr><th>acc_stderr</th><td>0.046056618647183814</td></tr><tr><th>acc_norm</th><td>0.3</td></tr><tr><th>acc_norm_stderr</th><td>0.046056618647183814</td></tr></table></td></tr><tr><th>hendrycksTest-clinical_knowledge</th><td><table><tr><th>acc</th><td>0.2641509433962264</td></tr><tr><th>acc_stderr</th><td>0.027134291628741713</td></tr><tr><th>acc_norm</th><td>0.2641509433962264</td></tr><tr><th>acc_norm_stderr</th><td>0.027134291628741713</td></tr></table></td></tr><tr><th>hendrycksTest-college_biology</th><td><table><tr><th>acc</th><td>0.2569444444444444</td></tr><tr><th>acc_stderr</th><td>0.03653946969442099</td></tr><tr><th>acc_norm</th><td>0.2569444444444444</td></tr><tr><th>acc_norm_stderr</th><td>0.03653946969442099</td></tr></table></td></tr><tr><th>hendrycksTest-college_chemistry</th><td><table><tr><th>acc</th><td>0.22</td></tr><tr><th>acc_stderr</th><td>0.041633319989322695</td></tr><tr><th>acc_norm</th><td>0.22</td></tr><tr><th>acc_norm_stderr</th><td>0.041633319989322695</td></tr></table></td></tr><tr><th>hendrycksTest-college_computer_science</th><td><table><tr><th>acc</th><td>0.26</td></tr><tr><th>acc_stderr</th><td>0.0440844002276808</td></tr><tr><th>acc_norm</th><td>0.26</td></tr><tr><th>acc_norm_stderr</th><td>0.0440844002276808</td></tr></table></td></tr><tr><th>hendrycksTest-college_mathematics</th><td><table><tr><th>acc</th><td>0.31</td></tr><tr><th>acc_stderr</th><td>0.04648231987117316</td></tr><tr><th>acc_norm</th><td>0.31</td></tr><tr><th>acc_norm_stderr</th><td>0.04648231987117316</td></tr></table></td></tr><tr><th>hendrycksTest-college_medicine</th><td><table><tr><th>acc</th><td>0.23121387283236994</td></tr><tr><th>acc_stderr</th><td>0.032147373020294696</td></tr><tr><th>acc_norm</th><td>0.23121387283236994</td></tr><tr><th>acc_norm_stderr</th><td>0.032147373020294696</td></tr></table></td></tr><tr><th>hendrycksTest-college_physics</th><td><table><tr><th>acc</th><td>0.27450980392156865</td></tr><tr><th>acc_stderr</th><td>0.04440521906179327</td></tr><tr><th>acc_norm</th><td>0.27450980392156865</td></tr><tr><th>acc_norm_stderr</th><td>0.04440521906179327</td></tr></table></td></tr><tr><th>hendrycksTest-computer_security</th><td><table><tr><th>acc</th><td>0.36</td></tr><tr><th>acc_stderr</th><td>0.048241815132442176</td></tr><tr><th>acc_norm</th><td>0.36</td></tr><tr><th>acc_norm_stderr</th><td>0.048241815132442176</td></tr></table></td></tr><tr><th>hendrycksTest-conceptual_physics</th><td><table><tr><th>acc</th><td>0.2765957446808511</td></tr><tr><th>acc_stderr</th><td>0.029241883869628837</td></tr><tr><th>acc_norm</th><td>0.2765957446808511</td></tr><tr><th>acc_norm_stderr</th><td>0.029241883869628837</td></tr></table></td></tr><tr><th>hendrycksTest-econometrics</th><td><table><tr><th>acc</th><td>0.2631578947368421</td></tr><tr><th>acc_stderr</th><td>0.04142439719489363</td></tr><tr><th>acc_norm</th><td>0.2631578947368421</td></tr><tr><th>acc_norm_stderr</th><td>0.04142439719489363</td></tr></table></td></tr><tr><th>hendrycksTest-electrical_engineering</th><td><table><tr><th>acc</th><td>0.20689655172413793</td></tr><tr><th>acc_stderr</th><td>0.03375672449560554</td></tr><tr><th>acc_norm</th><td>0.20689655172413793</td></tr><tr><th>acc_norm_stderr</th><td>0.03375672449560554</td></tr></table></td></tr><tr><th>hendrycksTest-elementary_mathematics</th><td><table><tr><th>acc</th><td>0.2698412698412698</td></tr><tr><th>acc_stderr</th><td>0.022860838309232072</td></tr><tr><th>acc_norm</th><td>0.2698412698412698</td></tr><tr><th>acc_norm_stderr</th><td>0.022860838309232072</td></tr></table></td></tr><tr><th>hendrycksTest-formal_logic</th><td><table><tr><th>acc</th><td>0.2619047619047619</td></tr><tr><th>acc_stderr</th><td>0.039325376803928704</td></tr><tr><th>acc_norm</th><td>0.2619047619047619</td></tr><tr><th>acc_norm_stderr</th><td>0.039325376803928704</td></tr></table></td></tr><tr><th>hendrycksTest-global_facts</th><td><table><tr><th>acc</th><td>0.35</td></tr><tr><th>acc_stderr</th><td>0.047937248544110196</td></tr><tr><th>acc_norm</th><td>0.35</td></tr><tr><th>acc_norm_stderr</th><td>0.047937248544110196</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_biology</th><td><table><tr><th>acc</th><td>0.24193548387096775</td></tr><tr><th>acc_stderr</th><td>0.0243625996930311</td></tr><tr><th>acc_norm</th><td>0.24193548387096775</td></tr><tr><th>acc_norm_stderr</th><td>0.0243625996930311</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_chemistry</th><td><table><tr><th>acc</th><td>0.28078817733990147</td></tr><tr><th>acc_stderr</th><td>0.0316185633535861</td></tr><tr><th>acc_norm</th><td>0.28078817733990147</td></tr><tr><th>acc_norm_stderr</th><td>0.0316185633535861</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_computer_science</th><td><table><tr><th>acc</th><td>0.33</td></tr><tr><th>acc_stderr</th><td>0.04725815626252605</td></tr><tr><th>acc_norm</th><td>0.33</td></tr><tr><th>acc_norm_stderr</th><td>0.04725815626252605</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_european_history</th><td><table><tr><th>acc</th><td>0.296969696969697</td></tr><tr><th>acc_stderr</th><td>0.03567969772268048</td></tr><tr><th>acc_norm</th><td>0.296969696969697</td></tr><tr><th>acc_norm_stderr</th><td>0.03567969772268048</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_geography</th><td><table><tr><th>acc</th><td>0.2878787878787879</td></tr><tr><th>acc_stderr</th><td>0.03225883512300993</td></tr><tr><th>acc_norm</th><td>0.2878787878787879</td></tr><tr><th>acc_norm_stderr</th><td>0.03225883512300993</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_government_and_politics</th><td><table><tr><th>acc</th><td>0.2538860103626943</td></tr><tr><th>acc_stderr</th><td>0.0314102478056532</td></tr><tr><th>acc_norm</th><td>0.2538860103626943</td></tr><tr><th>acc_norm_stderr</th><td>0.0314102478056532</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_macroeconomics</th><td><table><tr><th>acc</th><td>0.2743589743589744</td></tr><tr><th>acc_stderr</th><td>0.022622765767493207</td></tr><tr><th>acc_norm</th><td>0.2743589743589744</td></tr><tr><th>acc_norm_stderr</th><td>0.022622765767493207</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_mathematics</th><td><table><tr><th>acc</th><td>0.26296296296296295</td></tr><tr><th>acc_stderr</th><td>0.026842057873833706</td></tr><tr><th>acc_norm</th><td>0.26296296296296295</td></tr><tr><th>acc_norm_stderr</th><td>0.026842057873833706</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_microeconomics</th><td><table><tr><th>acc</th><td>0.2647058823529412</td></tr><tr><th>acc_stderr</th><td>0.028657491285071977</td></tr><tr><th>acc_norm</th><td>0.2647058823529412</td></tr><tr><th>acc_norm_stderr</th><td>0.028657491285071977</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_physics</th><td><table><tr><th>acc</th><td>0.304635761589404</td></tr><tr><th>acc_stderr</th><td>0.03757949922943343</td></tr><tr><th>acc_norm</th><td>0.304635761589404</td></tr><tr><th>acc_norm_stderr</th><td>0.03757949922943343</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_psychology</th><td><table><tr><th>acc</th><td>0.28623853211009176</td></tr><tr><th>acc_stderr</th><td>0.019379436628919968</td></tr><tr><th>acc_norm</th><td>0.28623853211009176</td></tr><tr><th>acc_norm_stderr</th><td>0.019379436628919968</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_statistics</th><td><table><tr><th>acc</th><td>0.25462962962962965</td></tr><tr><th>acc_stderr</th><td>0.02971127586000535</td></tr><tr><th>acc_norm</th><td>0.25462962962962965</td></tr><tr><th>acc_norm_stderr</th><td>0.02971127586000535</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_us_history</th><td><table><tr><th>acc</th><td>0.23039215686274508</td></tr><tr><th>acc_stderr</th><td>0.029554292605695053</td></tr><tr><th>acc_norm</th><td>0.23039215686274508</td></tr><tr><th>acc_norm_stderr</th><td>0.029554292605695053</td></tr></table></td></tr><tr><th>hendrycksTest-high_school_world_history</th><td><table><tr><th>acc</th><td>0.2869198312236287</td></tr><tr><th>acc_stderr</th><td>0.029443773022594693</td></tr><tr><th>acc_norm</th><td>0.2869198312236287</td></tr><tr><th>acc_norm_stderr</th><td>0.029443773022594693</td></tr></table></td></tr><tr><th>hendrycksTest-human_aging</th><td><table><tr><th>acc</th><td>0.3811659192825112</td></tr><tr><th>acc_stderr</th><td>0.03259625118416827</td></tr><tr><th>acc_norm</th><td>0.3811659192825112</td></tr><tr><th>acc_norm_stderr</th><td>0.03259625118416827</td></tr></table></td></tr><tr><th>hendrycksTest-human_sexuality</th><td><table><tr><th>acc</th><td>0.1984732824427481</td></tr><tr><th>acc_stderr</th><td>0.03498149385462472</td></tr><tr><th>acc_norm</th><td>0.1984732824427481</td></tr><tr><th>acc_norm_stderr</th><td>0.03498149385462472</td></tr></table></td></tr><tr><th>hendrycksTest-international_law</th><td><table><tr><th>acc</th><td>0.3884297520661157</td></tr><tr><th>acc_stderr</th><td>0.04449270350068382</td></tr><tr><th>acc_norm</th><td>0.3884297520661157</td></tr><tr><th>acc_norm_stderr</th><td>0.04449270350068382</td></tr></table></td></tr><tr><th>hendrycksTest-jurisprudence</th><td><table><tr><th>acc</th><td>0.23148148148148148</td></tr><tr><th>acc_stderr</th><td>0.04077494709252627</td></tr><tr><th>acc_norm</th><td>0.23148148148148148</td></tr><tr><th>acc_norm_stderr</th><td>0.04077494709252627</td></tr></table></td></tr><tr><th>hendrycksTest-logical_fallacies</th><td><table><tr><th>acc</th><td>0.2331288343558282</td></tr><tr><th>acc_stderr</th><td>0.03322015795776741</td></tr><tr><th>acc_norm</th><td>0.2331288343558282</td></tr><tr><th>acc_norm_stderr</th><td>0.03322015795776741</td></tr></table></td></tr><tr><th>hendrycksTest-machine_learning</th><td><table><tr><th>acc</th><td>0.21428571428571427</td></tr><tr><th>acc_stderr</th><td>0.03894641120044792</td></tr><tr><th>acc_norm</th><td>0.21428571428571427</td></tr><tr><th>acc_norm_stderr</th><td>0.03894641120044792</td></tr></table></td></tr><tr><th>hendrycksTest-management</th><td><table><tr><th>acc</th><td>0.3300970873786408</td></tr><tr><th>acc_stderr</th><td>0.04656147110012352</td></tr><tr><th>acc_norm</th><td>0.3300970873786408</td></tr><tr><th>acc_norm_stderr</th><td>0.04656147110012352</td></tr></table></td></tr><tr><th>hendrycksTest-marketing</th><td><table><tr><th>acc</th><td>0.2905982905982906</td></tr><tr><th>acc_stderr</th><td>0.029745048572674078</td></tr><tr><th>acc_norm</th><td>0.2905982905982906</td></tr><tr><th>acc_norm_stderr</th><td>0.029745048572674078</td></tr></table></td></tr><tr><th>hendrycksTest-medical_genetics</th><td><table><tr><th>acc</th><td>0.29</td></tr><tr><th>acc_stderr</th><td>0.04560480215720684</td></tr><tr><th>acc_norm</th><td>0.29</td></tr><tr><th>acc_norm_stderr</th><td>0.04560480215720684</td></tr></table></td></tr><tr><th>hendrycksTest-miscellaneous</th><td><table><tr><th>acc</th><td>0.31545338441890164</td></tr><tr><th>acc_stderr</th><td>0.016617501738763394</td></tr><tr><th>acc_norm</th><td>0.31545338441890164</td></tr><tr><th>acc_norm_stderr</th><td>0.016617501738763394</td></tr></table></td></tr><tr><th>hendrycksTest-moral_disputes</th><td><table><tr><th>acc</th><td>0.2861271676300578</td></tr><tr><th>acc_stderr</th><td>0.02433214677913413</td></tr><tr><th>acc_norm</th><td>0.2861271676300578</td></tr><tr><th>acc_norm_stderr</th><td>0.02433214677913413</td></tr></table></td></tr><tr><th>hendrycksTest-moral_scenarios</th><td><table><tr><th>acc</th><td>0.2122905027932961</td></tr><tr><th>acc_stderr</th><td>0.01367664468583173</td></tr><tr><th>acc_norm</th><td>0.2122905027932961</td></tr><tr><th>acc_norm_stderr</th><td>0.01367664468583173</td></tr></table></td></tr><tr><th>hendrycksTest-nutrition</th><td><table><tr><th>acc</th><td>0.2875816993464052</td></tr><tr><th>acc_stderr</th><td>0.02591780611714716</td></tr><tr><th>acc_norm</th><td>0.2875816993464052</td></tr><tr><th>acc_norm_stderr</th><td>0.02591780611714716</td></tr></table></td></tr><tr><th>hendrycksTest-philosophy</th><td><table><tr><th>acc</th><td>0.2765273311897106</td></tr><tr><th>acc_stderr</th><td>0.02540383297817961</td></tr><tr><th>acc_norm</th><td>0.2765273311897106</td></tr><tr><th>acc_norm_stderr</th><td>0.02540383297817961</td></tr></table></td></tr><tr><th>hendrycksTest-prehistory</th><td><table><tr><th>acc</th><td>0.3117283950617284</td></tr><tr><th>acc_stderr</th><td>0.025773111169630446</td></tr><tr><th>acc_norm</th><td>0.3117283950617284</td></tr><tr><th>acc_norm_stderr</th><td>0.025773111169630446</td></tr></table></td></tr><tr><th>hendrycksTest-professional_accounting</th><td><table><tr><th>acc</th><td>0.26595744680851063</td></tr><tr><th>acc_stderr</th><td>0.026358065698880592</td></tr><tr><th>acc_norm</th><td>0.26595744680851063</td></tr><tr><th>acc_norm_stderr</th><td>0.026358065698880592</td></tr></table></td></tr><tr><th>hendrycksTest-professional_law</th><td><table><tr><th>acc</th><td>0.25684485006518903</td></tr><tr><th>acc_stderr</th><td>0.011158455853098832</td></tr><tr><th>acc_norm</th><td>0.25684485006518903</td></tr><tr><th>acc_norm_stderr</th><td>0.011158455853098832</td></tr></table></td></tr><tr><th>hendrycksTest-professional_medicine</th><td><table><tr><th>acc</th><td>0.1801470588235294</td></tr><tr><th>acc_stderr</th><td>0.023345163616544855</td></tr><tr><th>acc_norm</th><td>0.1801470588235294</td></tr><tr><th>acc_norm_stderr</th><td>0.023345163616544855</td></tr></table></td></tr><tr><th>hendrycksTest-professional_psychology</th><td><table><tr><th>acc</th><td>0.27941176470588236</td></tr><tr><th>acc_stderr</th><td>0.018152871051538802</td></tr><tr><th>acc_norm</th><td>0.27941176470588236</td></tr><tr><th>acc_norm_stderr</th><td>0.018152871051538802</td></tr></table></td></tr><tr><th>hendrycksTest-public_relations</th><td><table><tr><th>acc</th><td>0.3090909090909091</td></tr><tr><th>acc_stderr</th><td>0.044262946482000985</td></tr><tr><th>acc_norm</th><td>0.3090909090909091</td></tr><tr><th>acc_norm_stderr</th><td>0.044262946482000985</td></tr></table></td></tr><tr><th>hendrycksTest-security_studies</th><td><table><tr><th>acc</th><td>0.32653061224489793</td></tr><tr><th>acc_stderr</th><td>0.030021056238440313</td></tr><tr><th>acc_norm</th><td>0.32653061224489793</td></tr><tr><th>acc_norm_stderr</th><td>0.030021056238440313</td></tr></table></td></tr><tr><th>hendrycksTest-sociology</th><td><table><tr><th>acc</th><td>0.25870646766169153</td></tr><tr><th>acc_stderr</th><td>0.030965903123573026</td></tr><tr><th>acc_norm</th><td>0.25870646766169153</td></tr><tr><th>acc_norm_stderr</th><td>0.030965903123573026</td></tr></table></td></tr><tr><th>hendrycksTest-us_foreign_policy</th><td><table><tr><th>acc</th><td>0.32</td></tr><tr><th>acc_stderr</th><td>0.04688261722621504</td></tr><tr><th>acc_norm</th><td>0.32</td></tr><tr><th>acc_norm_stderr</th><td>0.04688261722621504</td></tr></table></td></tr><tr><th>hendrycksTest-virology</th><td><table><tr><th>acc</th><td>0.30120481927710846</td></tr><tr><th>acc_stderr</th><td>0.0357160923005348</td></tr><tr><th>acc_norm</th><td>0.30120481927710846</td></tr><tr><th>acc_norm_stderr</th><td>0.0357160923005348</td></tr></table></td></tr><tr><th>hendrycksTest-world_religions</th><td><table><tr><th>acc</th><td>0.32748538011695905</td></tr><tr><th>acc_stderr</th><td>0.035993357714560276</td></tr><tr><th>acc_norm</th><td>0.32748538011695905</td></tr><tr><th>acc_norm_stderr</th><td>0.035993357714560276</td></tr></table></td></tr><tr><th>truthfulqa_mc</th><td><table><tr><th>mc1</th><td>0.2423500611995104</td></tr><tr><th>mc1_stderr</th><td>0.01500067437357034</td></tr><tr><th>mc2</th><td>0.3859757929597962</td></tr><tr><th>mc2_stderr</th><td>0.013898628036488968</td></tr></table></td></tr></table>