mainsafetensorsbf16Public
bluecolorv2/teutonic-5ef1xygw-ultradata-capability-adapter-step40
sha256:584c71c1373206f8e24acc22315ed12e39e83ea31665c9b1306019fcfaea62ca·Indexed 17d ago
Parameters
8.6B
Total size
16.0 GB
Files
8
Quantization
BF16
No README on this version
Push a README.md with the model files to see it rendered here.
Model architecture
config.json- Architecture
- QuasarForCausalLM
- Model type
- quasar_text
- Hidden size
- 4,096
- Layers
- 32
- Attention heads
- 16 (4 kv)
- FFN size
- 12,288
- Vocab size
- 248,320
- Context window
- 2048K
Files
8 itemsmodel.safetensors
7614e2738b59
16.0 GB
safetensors
tokenizer.json
87a7830d63fc
19.1 MB
modeling_qwen3_5.py
a9e25ab03df1
104.7 KB
paired_eval_result.json
a7f3aea7b798
31.9 KB
configuration_qwen3_5.py
6e5d4d9ef7b2
14.4 KB
config.json
edc73f163432
1.9 KB
tokenizer_config.json
930c80792230
1.1 KB
generation_config.json
a7c7a2fe3294
215 B