Downloads · 30 days
18
3% of all-time downloads
ABX-AI_-_Silver-Sun-11B-gguf
RichardErkhov/ABX-AI_-_Silver-Sun-11B-gguf
ABX-AI_-_Silver-Sun-11B-gguf is a machine learning model from RichardErkhov. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
All-time downloads
524
Public
Repo size
122 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf141 GB · 100%
At a glance
- Access
- Public
- Created
- May 24, 2024
- Updated
- May 24, 2024
- SHA
- ed470ffb
Model card
From the Hugging Face model README
Quantization made by Richard Erkhov.
Silver-Sun-11B - GGUF
- Model creator: https://huggingface.co/ABX-AI/
- Original model: https://huggingface.co/ABX-AI/Silver-Sun-11B/
| Name | Quant method | Size |
|---|---|---|
| Silver-Sun-11B.Q2_K.gguf | Q2_K | 3.73GB |
| Silver-Sun-11B.IQ3_XS.gguf | IQ3_XS | 4.14GB |
| Silver-Sun-11B.IQ3_S.gguf | IQ3_S | 4.37GB |
| Silver-Sun-11B.Q3_K_S.gguf | Q3_K_S | 4.34GB |
| Silver-Sun-11B.IQ3_M.gguf | IQ3_M | 4.51GB |
| Silver-Sun-11B.Q3_K.gguf | Q3_K | 4.84GB |
| Silver-Sun-11B.Q3_K_M.gguf | Q3_K_M | 4.84GB |
| Silver-Sun-11B.Q3_K_L.gguf | Q3_K_L | 5.26GB |
| Silver-Sun-11B.IQ4_XS.gguf | IQ4_XS | 5.43GB |
| Silver-Sun-11B.Q4_0.gguf | Q4_0 | 5.66GB |
| Silver-Sun-11B.IQ4_NL.gguf | IQ4_NL | 5.72GB |
| Silver-Sun-11B.Q4_K_S.gguf | Q4_K_S | 5.7GB |
| Silver-Sun-11B.Q4_K.gguf | Q4_K | 6.02GB |
| Silver-Sun-11B.Q4_K_M.gguf | Q4_K_M | 6.02GB |
| Silver-Sun-11B.Q4_1.gguf | Q4_1 | 6.27GB |
| Silver-Sun-11B.Q5_0.gguf | Q5_0 | 6.89GB |
| Silver-Sun-11B.Q5_K_S.gguf | Q5_K_S | 6.89GB |
| Silver-Sun-11B.Q5_K.gguf | Q5_K | 7.08GB |
| Silver-Sun-11B.Q5_K_M.gguf | Q5_K_M | 7.08GB |
| Silver-Sun-11B.Q5_1.gguf | Q5_1 | 7.51GB |
| Silver-Sun-11B.Q6_K.gguf | Q6_K | 8.2GB |
| Silver-Sun-11B.Q8_0.gguf | Q8_0 | 10.62GB |
Original model description:
license: other tags:
- merge
- llama
- not-for-all-audiences
- mergekit
Silver-Sun-11B
I'd like to experiment more with merging 11B, hopefully adding more options of this weight class. This model is good at writing and reasoning, with a preference for more profane NSFW language when the appropriate cards are used. I've been having fun with it so far, although at times it can be a bit blunt, although some may prefer that. It's also highly uncensored.
Works best with Alpaca instruction presets.
Merge Details
This is a merge of pre-trained language models created using mergekit.
Merge Method
This model was merged using the SLERP merge method.
Models Merged
The following models were included in the merge:
- ABX-AI/Solstice-FKL-11B
[!NOTE] A mixture of Sao10K/Solstice-11B-v1 and saishf/Fimbulvetr-Kuro-Lotus-10.7B
OpenLLM Eval Results
Detailed Results + Failed GSM8K
[!NOTE] I had to remove GSM8K from the results and manually re-average the rest. GSM8K failed due to some issue with formatting, which is not experienced during practical usage. By removing the GSM8K score, the average is VERY close to upstage/SOLAR-10.7B-v1.0 (74.20), which would make sense. Feel free to ignore the actual average and use the other scores individually for reference.
| Metric | Value |
|---|---|
| Avg. | 74.13 |
| AI2 Reasoning Challenge (25-Shot) | 69.80 |
| HellaSwag (10-Shot) | 87.91 |
| MMLU (5-Shot) | 66.90 |
| TruthfulQA (0-shot) | 61.89 |
| Winogrande (5-shot) | 84.14 |
Configuration
The following YAML configuration was used to produce this model:
slices:
- sources:
- model: ABX-AI/Solstice-FKL-11B
layer_range: [0, 48]
- model: Himitsui/Kaiju-11B
layer_range: [0, 48]
merge_method: slerp
base_model: ABX-AI/Solstice-FKL-11B
parameters:
t:
- filter: self_attn
value: [0.6, 0.7, 0.8, 0.9, 1]
- filter: mlp
value: [0.4, 0.3, 0.2, 0.1, 0]
- value: 0.5
dtype: bfloat16
- Created
- May 24, 2024
- Updated
- May 24, 2024