Downloads · 30 days
102
5% of all-time downloads
RichardErkhov/PY007_-_EasyContext-256K-danube2-1.8b-gguf
PY007_-_EasyContext-256K-danube2-1.8b-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.
Downloads · 30 days
102
5% of all-time downloads
All-time downloads
2K
Public
Repo size
18.6 GB
Likes
0
Public
Click a slice to open those files.
.gguf21.9 GB · 100%
From the Hugging Face model README
Quantization made by Richard Erkhov.
EasyContext-256K-danube2-1.8b - GGUF
| Name | Quant method | Size |
|---|---|---|
| EasyContext-256K-danube2-1.8b.Q2_K.gguf | Q2_K | 0.66GB |
| EasyContext-256K-danube2-1.8b.Q3_K_S.gguf | Q3_K_S | 0.76GB |
| EasyContext-256K-danube2-1.8b.Q3_K.gguf | Q3_K | 0.84GB |
| EasyContext-256K-danube2-1.8b.Q3_K_M.gguf | Q3_K_M | 0.84GB |
| EasyContext-256K-danube2-1.8b.Q3_K_L.gguf | Q3_K_L | 0.91GB |
| EasyContext-256K-danube2-1.8b.IQ4_XS.gguf | IQ4_XS | 0.94GB |
| EasyContext-256K-danube2-1.8b.Q4_0.gguf | Q4_0 | 0.98GB |
| EasyContext-256K-danube2-1.8b.IQ4_NL.gguf | IQ4_NL | 0.99GB |
| EasyContext-256K-danube2-1.8b.Q4_K_S.gguf | Q4_K_S | 0.99GB |
| EasyContext-256K-danube2-1.8b.Q4_K.gguf | Q4_K | 1.04GB |
| EasyContext-256K-danube2-1.8b.Q4_K_M.gguf | Q4_K_M | 1.04GB |
| EasyContext-256K-danube2-1.8b.Q4_1.gguf | Q4_1 | 1.08GB |
| EasyContext-256K-danube2-1.8b.Q5_0.gguf | Q5_0 | 1.18GB |
| EasyContext-256K-danube2-1.8b.Q5_K_S.gguf | Q5_K_S | 1.18GB |
| EasyContext-256K-danube2-1.8b.Q5_K.gguf | Q5_K | 1.21GB |
| EasyContext-256K-danube2-1.8b.Q5_K_M.gguf | Q5_K_M | 1.21GB |
| EasyContext-256K-danube2-1.8b.Q5_1.gguf | Q5_1 | 1.29GB |
| EasyContext-256K-danube2-1.8b.Q6_K.gguf | Q6_K | 1.4GB |
| EasyContext-256K-danube2-1.8b.Q8_0.gguf | Q8_0 | 1.81GB |
Original model description:
Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.
This is a context-extrapolated base model. It has not been instruct-finetuned.
This model is finetuned from h2oai/h2o-danube2-1.8b-base with EasyContext on context length 256K. Note that I keep max_position_embeddings in config.json to 4096 because HF llama will create 2D causal mask during initialization. If it is set to 256K GPU will just OOM. You can surely use this model with context length longer than 4096.
<p align="center"> <img src="./heatmap.png" width="800"> </p>