trollek commited on
Commit
8526f56
·
verified ·
1 Parent(s): 7847e2a

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -18,10 +18,12 @@ base_model: trollek/danube2-1.8b-SlimOrcaDedup
18
  dtype: bfloat16
19
  ```
20
 
 
 
21
  ### LLama-Factory config
22
  ```yaml
23
  ### model
24
- model_name_or_path: /home/trolle/Documents/Projects/trollek/danube2/thinking-base-chatml
25
 
26
  ### method
27
  stage: sft
@@ -68,5 +70,3 @@ per_device_eval_batch_size: 1
68
  eval_strategy: steps
69
  eval_steps: 100
70
  ```
71
-
72
- The eval went up after 1 epoch so I stopped and merged that.
 
18
  dtype: bfloat16
19
  ```
20
 
21
+ The thought tokens were added to the merged model before training a LoRA.
22
+
23
  ### LLama-Factory config
24
  ```yaml
25
  ### model
26
+ model_name_or_path: merged-base-chatml
27
 
28
  ### method
29
  stage: sft
 
70
  eval_strategy: steps
71
  eval_steps: 100
72
  ```