Update README.md
Browse files
README.md
CHANGED
@@ -2,11 +2,12 @@
|
|
2 |
library_name: transformers
|
3 |
tags: []
|
4 |
---
|
5 |
-
Input, output embedding weights are possibly not tied -
|
6 |
-
|
7 |
Some weights of LlamaForCausalLM were not initialized from the model checkpoint at ./llama3-pa/llama3-32kpa-emb-init-weight-tied and are newly initialized: ['lm_head.weight']
|
8 |
|
9 |
-
used -
|
|
|
10 |
source_model.lm_head.weight.data = source_model.model.embed_tokens.weight.data
|
11 |
source_model.lm_head.weight = source_model.model.embed_tokens.weight
|
12 |
-
source_model.tie_weights()
|
|
|
|
2 |
library_name: transformers
|
3 |
tags: []
|
4 |
---
|
5 |
+
### Input, output embedding weights are possibly not tied -
|
|
|
6 |
Some weights of LlamaForCausalLM were not initialized from the model checkpoint at ./llama3-pa/llama3-32kpa-emb-init-weight-tied and are newly initialized: ['lm_head.weight']
|
7 |
|
8 |
+
### used -
|
9 |
+
```
|
10 |
source_model.lm_head.weight.data = source_model.model.embed_tokens.weight.data
|
11 |
source_model.lm_head.weight = source_model.model.embed_tokens.weight
|
12 |
+
source_model.tie_weights()
|
13 |
+
```
|