IceLemonTeaRP-32k-7b
This is a merge of pre-trained language models created using mergekit.
I would suggest to play with rope_theta in config.json to set between 40000-100000.
Merge Details
Cooked merge from fresh ingredients to fix icefog72/IceTeaRP-7b repetition problems.
Prompt template: Alpaca, maybe ChatML
- measurement.json for quanting exl2 included.
- IceLemonTeaRP-32k-7b-4.0bpw-h6-exl2
- IceLemonTeaRP-32k-7b-4.2bpw-h6-exl2
- IceLemonTeaRP-32k-7b-6.5bpw-h6-exl2
- IceLemonTeaRP-32k-7b-8.0bpw-h6-exl2
Thanks for
Merge Method
This model was merged using the SLERP merge method.
Models Merged
The following models were included in the merge:
- icefog72/Kunokukulemonchini-32k-7b
- icefog72/Mixtral_AI_Cyber_3.m1-BigL
How to download From the command line
I recommend using the huggingface-hub
Python library:
pip3 install huggingface-hub
To download the main
branch to a folder called IceLemonTeaRP-32k-7b
:
mkdir IceLemonTeaRP-32k-7b
huggingface-cli download icefog72/IceLemonTeaRP-32k-7b --local-dir IceLemonTeaRP-32k-7b --local-dir-use-symlinks False
More advanced huggingface-cli download usage
If you remove the --local-dir-use-symlinks False
parameter, the files will instead be stored in the central Hugging Face cache directory (default location on Linux is: ~/.cache/huggingface
), and symlinks will be added to the specified --local-dir
, pointing to their real location in the cache. This allows for interrupted downloads to be resumed, and allows you to quickly clone the repo to multiple places on disk without triggering a download again. The downside, and the reason why I don't list that as the default option, is that the files are then hidden away in a cache folder and it's harder to know where your disk space is being used, and to clear it up if/when you want to remove a download model.
The cache location can be changed with the HF_HOME
environment variable, and/or the --cache-dir
parameter to huggingface-cli
.
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
:
mkdir FOLDERNAME
HF_HUB_ENABLE_HF_TRANSFER=1 huggingface-cli download MODEL --local-dir FOLDERNAME --local-dir-use-symlinks False
Windows Command Line users: You can set the environment variable by running set HF_HUB_ENABLE_HF_TRANSFER=1
before the download command.
The following YAML configuration was used to produce this model:
slices:
- sources:
- model: Mixtral_AI_Cyber_3.m1-BigL
layer_range: [0, 32]
- model: Kunokukulemonchini-32k-7b
layer_range: [0, 32]
merge_method: slerp
base_model: Kunokukulemonchini-32k-7b
parameters:
t:
- filter: self_attn
value: [0, 0.5, 0.3, 0.7, 1]
- filter: mlp
value: [1, 0.5, 0.7, 0.3, 0]
- value: 0.5
dtype: float16
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 70.43 |
AI2 Reasoning Challenge (25-Shot) | 67.66 |
HellaSwag (10-Shot) | 86.53 |
MMLU (5-Shot) | 64.51 |
TruthfulQA (0-shot) | 61.76 |
Winogrande (5-shot) | 79.72 |
GSM8k (5-shot) | 62.40 |
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
Metric | Value |
---|---|
Avg. | 21.27 |
IFEval (0-Shot) | 52.12 |
BBH (3-Shot) | 30.14 |
MATH Lvl 5 (4-Shot) | 4.83 |
GPQA (0-shot) | 5.37 |
MuSR (0-shot) | 12.20 |
MMLU-PRO (5-shot) | 22.97 |
- Downloads last month
- 98
Model tree for icefog72/IceLemonTeaRP-32k-7b
Spaces using icefog72/IceLemonTeaRP-32k-7b 5
Collection including icefog72/IceLemonTeaRP-32k-7b
Evaluation results
- normalized accuracy on AI2 Reasoning Challenge (25-Shot)test set Open LLM Leaderboard67.660
- normalized accuracy on HellaSwag (10-Shot)validation set Open LLM Leaderboard86.530
- accuracy on MMLU (5-Shot)test set Open LLM Leaderboard64.510
- mc2 on TruthfulQA (0-shot)validation set Open LLM Leaderboard61.760
- accuracy on Winogrande (5-shot)validation set Open LLM Leaderboard79.720
- accuracy on GSM8k (5-shot)test set Open LLM Leaderboard62.400
- strict accuracy on IFEval (0-Shot)Open LLM Leaderboard52.120
- normalized accuracy on BBH (3-Shot)Open LLM Leaderboard30.140
- exact match on MATH Lvl 5 (4-Shot)Open LLM Leaderboard4.830
- acc_norm on GPQA (0-shot)Open LLM Leaderboard5.370