tamang0000 commited on
Commit
406f23f
1 Parent(s): 4f2e0b0

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +29 -3
README.md CHANGED
@@ -11,12 +11,38 @@ tags:
11
  - trl
12
  ---
13
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
14
  # Uploaded model
15
 
16
  - **Developed by:** tamang0000
17
  - **License:** apache-2.0
18
  - **Finetuned from model :** unsloth/qwen2-0.5b-bnb-4bit
19
 
20
- This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.
21
-
22
- [<img src="https://raw.githubusercontent.com/unslothai/unsloth/main/images/unsloth%20made%20with%20love.png" width="200"/>](https://github.com/unslothai/unsloth)
 
11
  - trl
12
  ---
13
 
14
+ Start: কম্পিউটাৰ
15
+
16
+ Before:
17
+ কম্পিউটাৰ কোনো স্বাস্থ্য করা হয় না। একটি স্বাস্থ্য করা
18
+ কম্পিউটাৰ কোনো স্বাস্থ্য করা হয় না। একটি স্বাস্থ্য করা
19
+
20
+
21
+ After:
22
+ কম্পিউটাৰ পৰাপৰ পৰাপৰ পৰাপৰ পৰাপৰ পৰাপৰ পৰাপৰ প
23
+ কম্পিউটাৰ পৰাপৰ পৰাপৰ পৰাপৰ পৰাপৰ পৰাপৰ পৰাপৰ প
24
+ ```
25
+ max_seq_length = 512
26
+ per_device_train_batch_size = 2,
27
+ gradient_accumulation_steps = 4,
28
+ warmup_steps = 2,
29
+ max_steps = 10,
30
+ learning_rate = 0.0005,
31
+ fp16 = not torch.cuda.is_bf16_supported(),
32
+ bf16 = torch.cuda.is_bf16_supported(),
33
+ logging_steps = 1,
34
+ optim = "adamw_8bit",
35
+ weight_decay = 0.01,
36
+ lr_scheduler_type = "linear",
37
+ seed = 3407,
38
+ output_dir = "outputs",
39
+ ```
40
+
41
+ ---
42
  # Uploaded model
43
 
44
  - **Developed by:** tamang0000
45
  - **License:** apache-2.0
46
  - **Finetuned from model :** unsloth/qwen2-0.5b-bnb-4bit
47
 
48
+ This qwen2 model was trained 2x faster with [Unsloth](https://github.com/unslothai/unsloth) and Huggingface's TRL library.