KotshinZ commited on
Commit
eabe392
·
verified ·
1 Parent(s): 2840f89

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -0
README.md CHANGED
@@ -15,6 +15,7 @@ licence: license
15
 
16
  This model is a fine-tuned version of [KotshinZ/gpt2-RMT-7-mem512](https://huggingface.co/KotshinZ/gpt2-RMT-7-mem512) on the [HuggingFaceFW/fineweb-edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) dataset.
17
  It has been trained using [TRL](https://github.com/huggingface/trl).
 
18
 
19
  ## Quick start
20
 
@@ -33,6 +34,7 @@ print(output["generated_text"])
33
 
34
 
35
  This model was trained with SFT.
 
36
 
37
  ### Framework versions
38
 
 
15
 
16
  This model is a fine-tuned version of [KotshinZ/gpt2-RMT-7-mem512](https://huggingface.co/KotshinZ/gpt2-RMT-7-mem512) on the [HuggingFaceFW/fineweb-edu](https://huggingface.co/datasets/HuggingFaceFW/fineweb-edu) dataset.
17
  It has been trained using [TRL](https://github.com/huggingface/trl).
18
+ For use this model. You need clone [KotShinZ/Recurrent-Memory-Transformer_PreTrained.git](https://github.com/KotShinZ/Recurrent-Memory-Transformer_PreTrained)
19
 
20
  ## Quick start
21
 
 
34
 
35
 
36
  This model was trained with SFT.
37
+ This model memory_size = 512, n_backward = 8.
38
 
39
  ### Framework versions
40