Update README.md
Browse files
README.md
CHANGED
|
@@ -47,6 +47,7 @@ Because it's a good idea to make tiny LLMs. Some people already did with [MicroL
|
|
| 47 |
* **Not Instruction-Tuned:** It's only a base model, so it only completes text.
|
| 48 |
* **English-Only:** It's trained on English data (FineWeb), it's NOT multilingual.
|
| 49 |
* **Not a Standard Model:** It's NOT a Qwen/Llama/GPT model. Standard Transformers can't recognize this!
|
|
|
|
| 50 |
|
| 51 |
# Some other details
|
| 52 |
* It's trained on 80 million tokens of [FineWeb](https://huggingface.co/datasets/HuggingFaceFW/fineweb) (CC-MAIN-2025-26 snapshot), and the knowledge cutoff is June 2025.
|
|
|
|
| 47 |
* **Not Instruction-Tuned:** It's only a base model, so it only completes text.
|
| 48 |
* **English-Only:** It's trained on English data (FineWeb), it's NOT multilingual.
|
| 49 |
* **Not a Standard Model:** It's NOT a Qwen/Llama/GPT model. Standard Transformers can't recognize this!
|
| 50 |
+
* **Preview:** This is a preview version, it generates gibberish often. CinnabarLM 1 will solve this with Llama.
|
| 51 |
|
| 52 |
# Some other details
|
| 53 |
* It's trained on 80 million tokens of [FineWeb](https://huggingface.co/datasets/HuggingFaceFW/fineweb) (CC-MAIN-2025-26 snapshot), and the knowledge cutoff is June 2025.
|