docs: update model card
Browse files
README.md
CHANGED
|
@@ -19,7 +19,6 @@ Iridium is a 72B parameter language model created through a merge of Qwen2-72B-I
|
|
| 19 |
|
| 20 |
## Features
|
| 21 |
- 72 billion parameters
|
| 22 |
-
- Comes in 1,043 individual safetensor files
|
| 23 |
- Combines Magnum prose with Calam smarts
|
| 24 |
|
| 25 |
## Technical Specifications
|
|
@@ -28,14 +27,14 @@ Iridium is a 72B parameter language model created through a merge of Qwen2-72B-I
|
|
| 28 |
- `Qwen2ForCasualLM`
|
| 29 |
- Models: Qwen2-72B-Instruct (base), calme2.1-72b, magnum-72b-v1
|
| 30 |
- Merged layers: 80
|
| 31 |
-
- Total tensors:
|
| 32 |
- Context length: 128k
|
| 33 |
|
| 34 |
### Tensor Distribution
|
| 35 |
- Attention layers: 560 files
|
| 36 |
- MLP layers: 240 files
|
| 37 |
- Layer norms: 160 files
|
| 38 |
-
- Miscellaneous (embeddings, output):
|
| 39 |
|
| 40 |
### Merging
|
| 41 |
Custom script utilizing safetensors library.
|
|
|
|
| 19 |
|
| 20 |
## Features
|
| 21 |
- 72 billion parameters
|
|
|
|
| 22 |
- Combines Magnum prose with Calam smarts
|
| 23 |
|
| 24 |
## Technical Specifications
|
|
|
|
| 27 |
- `Qwen2ForCasualLM`
|
| 28 |
- Models: Qwen2-72B-Instruct (base), calme2.1-72b, magnum-72b-v1
|
| 29 |
- Merged layers: 80
|
| 30 |
+
- Total tensors: 963
|
| 31 |
- Context length: 128k
|
| 32 |
|
| 33 |
### Tensor Distribution
|
| 34 |
- Attention layers: 560 files
|
| 35 |
- MLP layers: 240 files
|
| 36 |
- Layer norms: 160 files
|
| 37 |
+
- Miscellaneous (embeddings, output): 3 files
|
| 38 |
|
| 39 |
### Merging
|
| 40 |
Custom script utilizing safetensors library.
|