base_model: jondurbin/bagel-dpo-20b-v04
datasets:
- ai2_arc
- allenai/ultrafeedback_binarized_cleaned
- argilla/distilabel-intel-orca-dpo-pairs
- jondurbin/airoboros-3.2
- codeparrot/apps
- facebook/belebele
- bluemoon-fandom-1-1-rp-cleaned
- boolq
- camel-ai/biology
- camel-ai/chemistry
- camel-ai/math
- camel-ai/physics
- jondurbin/contextual-dpo-v0.1
- jondurbin/gutenberg-dpo-v0.1
- jondurbin/py-dpo-v0.1
- jondurbin/truthy-dpo-v0.1
- LDJnr/Capybara
- jondurbin/cinematika-v0.1
- WizardLM/WizardLM_evol_instruct_70k
- glaiveai/glaive-function-calling-v2
- jondurbin/gutenberg-dpo-v0.1
- grimulkan/LimaRP-augmented
- lmsys/lmsys-chat-1m
- ParisNeo/lollms_aware_dataset
- TIGER-Lab/MathInstruct
- Muennighoff/natural-instructions
- openbookqa
- kingbri/PIPPA-shareGPT
- piqa
- Vezora/Tested-22k-Python-Alpaca
- ropes
- cakiki/rosetta-code
- Open-Orca/SlimOrca
- b-mc2/sql-create-context
- squad_v2
- mattpscott/airoboros-summarization
- migtissera/Synthia-v1.3
- unalignment/toxic-dpo-v0.2
- WhiteRabbitNeo/WRN-Chapter-1
- WhiteRabbitNeo/WRN-Chapter-2
- winogrande
language:
- en
library_name: transformers
license: other
license_link: https://huggingface.co/internlm/internlm2-20b#open-source-license
license_name: internlm2-20b
quantized_by: mradermacher
About
weighted/imatrix quants of https://huggingface.co/jondurbin/bagel-dpo-20b-v04
This uses only 95k tokens of my standard set, as the model overflowed with more.
static quants are available at https://huggingface.co/mradermacher/bagel-dpo-20b-v04-GGUF
Usage
If you are unsure how to use GGUF files, refer to one of TheBloke's READMEs for more details, including on how to concatenate multi-part files.
Provided Quants
(sorted by size, not necessarily quality. IQ-quants are often preferable over similar sized non-IQ quants)
| Link | Type | Size/GB | Notes |
|---|---|---|---|
| GGUF | i1-IQ1_S | 5.4 | for the desperate |
| GGUF | i1-IQ1_M | 5.8 | mostly desperate |
| GGUF | i1-IQ2_XXS | 6.4 | |
| GGUF | i1-IQ2_XS | 6.9 | |
| GGUF | i1-IQ2_S | 7.3 | |
| GGUF | i1-IQ2_M | 7.8 | |
| GGUF | i1-Q2_K | 8.3 | IQ3_XXS probably better |
| GGUF | i1-IQ3_XXS | 8.7 | lower quality |
| GGUF | i1-IQ3_XS | 9.1 | |
| GGUF | i1-Q3_K_S | 9.5 | IQ3_XS probably better |
| GGUF | i1-IQ3_S | 9.6 | beats Q3_K* |
| GGUF | i1-IQ3_M | 9.9 | |
| GGUF | i1-Q3_K_M | 10.5 | IQ3_S probably better |
| GGUF | i1-Q3_K_L | 11.3 | IQ3_M probably better |
| GGUF | i1-IQ4_XS | 11.5 | |
| GGUF | i1-Q4_0 | 12.1 | fast, low quality |
| GGUF | i1-Q4_K_S | 12.2 | optimal size/speed/quality |
| GGUF | i1-Q4_K_M | 12.8 | fast, recommended |
| GGUF | i1-Q5_K_S | 14.5 | |
| GGUF | i1-Q5_K_M | 14.8 | |
| GGUF | i1-Q6_K | 17.1 | practically like static Q6_K |
Here is a handy graph by ikawrakow comparing some lower-quality quant types (lower is better):
And here are Artefact2's thoughts on the matter: https://gist.github.com/Artefact2/b5f810600771265fc1e39442288e8ec9
FAQ / Model Request
See https://huggingface.co/mradermacher/model_requests for some answers to questions you might have and/or if you want some other model quantized.
Thanks
I thank my company, nethype GmbH, for letting me use its servers and providing upgrades to my workstation to enable this work in my free time.