Update README.md
Browse files
README.md
CHANGED
@@ -20,7 +20,8 @@ wizardphind-coder-passthrough-39B is a merge of the following models using [merg
|
|
20 |
wizardphind-coder-passthrough-39B is an experimental model combining the deepseek-33B and codellama-34B models.
|
21 |
I expect the model to become much better when trained further on coding specific tasks.
|
22 |
|
23 |
-
Since deepseek & the codellama models have different sized tensors for their MLP/Attention layers,
|
|
|
24 |
|
25 |
|
26 |
## 🧩 Configuration
|
|
|
20 |
wizardphind-coder-passthrough-39B is an experimental model combining the deepseek-33B and codellama-34B models.
|
21 |
I expect the model to become much better when trained further on coding specific tasks.
|
22 |
|
23 |
+
Since deepseek & the codellama models have different sized tensors for their MLP/Attention layers,
|
24 |
+
this model will be initialized with empty layers and will need to be fine-tuned futher.
|
25 |
|
26 |
|
27 |
## 🧩 Configuration
|