Upload README.md with huggingface_hub
Browse files
README.md
CHANGED
|
@@ -3,7 +3,6 @@ license: apache-2.0
|
|
| 3 |
pipeline_tag: text-generation
|
| 4 |
tags:
|
| 5 |
- base-model
|
| 6 |
-
- qwen3
|
| 7 |
language:
|
| 8 |
- en
|
| 9 |
---
|
|
@@ -23,7 +22,7 @@ Named after *Prochlorococcus*, the tiniest and most oxygen-producing photosynthe
|
|
| 23 |
| Vocab | 50,261 (GPT-2 BPE + 4 control tokens) |
|
| 24 |
| Context | 1024 |
|
| 25 |
| Precision | FP16 |
|
| 26 |
-
| Format | Safetensors,
|
| 27 |
|
| 28 |
## Pretraining
|
| 29 |
|
|
|
|
| 3 |
pipeline_tag: text-generation
|
| 4 |
tags:
|
| 5 |
- base-model
|
|
|
|
| 6 |
language:
|
| 7 |
- en
|
| 8 |
---
|
|
|
|
| 22 |
| Vocab | 50,261 (GPT-2 BPE + 4 control tokens) |
|
| 23 |
| Context | 1024 |
|
| 24 |
| Precision | FP16 |
|
| 25 |
+
| Format | Safetensors, loads with `Qwen3ForCausalLM` via `use_qk_norm=True` (architecture-compatible; trained from scratch by CyanoAI, not derived from Qwen) |
|
| 26 |
|
| 27 |
## Pretraining
|
| 28 |
|