Add README
Browse files
README.md
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
tags:
|
| 4 |
+
- nova-1
|
| 5 |
+
- language-model
|
| 6 |
+
- phase-1
|
| 7 |
+
---
|
| 8 |
+
|
| 9 |
+
# Nova-1 Large — Phase 1 Checkpoint
|
| 10 |
+
|
| 11 |
+
Phase 1 pretrained weights for Nova-1 Large.
|
| 12 |
+
|
| 13 |
+
## Files
|
| 14 |
+
|
| 15 |
+
```
|
| 16 |
+
nova1/
|
| 17 |
+
├── nova1_final_s0011211.safetensors # model weights
|
| 18 |
+
└── nova1_final_s0011211_config.json # architecture config
|
| 19 |
+
```
|
| 20 |
+
|
| 21 |
+
## Architecture
|
| 22 |
+
|
| 23 |
+
| Param | Value |
|
| 24 |
+
|---|---|
|
| 25 |
+
| d_model | 2560 |
|
| 26 |
+
| n_layers | 32 |
|
| 27 |
+
| n_heads | 20 |
|
| 28 |
+
| n_kv_heads | 10 |
|
| 29 |
+
| max_len | 2048 |
|
| 30 |
+
| vocab_size | 50304 |
|
| 31 |
+
|
| 32 |
+
Trained by SmilyAI Labs.
|