Update README.md
Browse files
README.md
CHANGED
|
@@ -12,6 +12,16 @@ tags:
|
|
| 12 |
This model was converted to GGUF format from [`SicariusSicariiStuff/Impish_Magic_24B`](https://huggingface.co/SicariusSicariiStuff/Impish_Magic_24B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
| 13 |
Refer to the [original model card](https://huggingface.co/SicariusSicariiStuff/Impish_Magic_24B) for more details on the model.
|
| 14 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
## Use with llama.cpp
|
| 16 |
Install llama.cpp through brew (works on Mac and Linux)
|
| 17 |
|
|
|
|
| 12 |
This model was converted to GGUF format from [`SicariusSicariiStuff/Impish_Magic_24B`](https://huggingface.co/SicariusSicariiStuff/Impish_Magic_24B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
| 13 |
Refer to the [original model card](https://huggingface.co/SicariusSicariiStuff/Impish_Magic_24B) for more details on the model.
|
| 14 |
|
| 15 |
+
---
|
| 16 |
+
This model is based on mistralai/Magistral-Small-2506 so naturally it's named Impish_Magic. Truly excellent size, it's been tested on a laptop with 16GB gpu and it works quite fast (4090m).
|
| 17 |
+
|
| 18 |
+
This model went "full" fine-tune over 100m unique tokens. Why "full"?
|
| 19 |
+
|
| 20 |
+
Specific areas in the model have been tuned to attempt to change the
|
| 21 |
+
vocabulary usage, while keeping as much intelligence as possible. So
|
| 22 |
+
this is definitely not a LoRA, but also not exactly a proper full finetune, but rather something in-between.
|
| 23 |
+
|
| 24 |
+
---
|
| 25 |
## Use with llama.cpp
|
| 26 |
Install llama.cpp through brew (works on Mac and Linux)
|
| 27 |
|