Text Generation
Transformers
Safetensors
English
llama
conversational
text-generation-inference
lucifertrj commited on
Commit
588bfaf
·
verified ·
1 Parent(s): a3ba5f8

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +15 -0
README.md CHANGED
@@ -26,6 +26,21 @@ import torch
26
  import transformers
27
  from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments,BitsAndBytesConfig
28
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
29
  prompt = f"""### Instruction:
30
  Below is an instruction that describes a task. Write a response that appropriately completes the request.
31
 
 
26
  import transformers
27
  from transformers import AutoModelForCausalLM, AutoTokenizer, TrainingArguments,BitsAndBytesConfig
28
 
29
+ bnb_config = BitsAndBytesConfig(
30
+ load_in_4bit=True,
31
+ bnb_4bit_use_double_quant=True,
32
+ bnb_4bit_quant_type="nf4",
33
+ bnb_4bit_compute_dtype=torch.bfloat16
34
+ )
35
+
36
+ model = "aiplanet/panda-coder-13B"
37
+
38
+ base_model = AutoModelForCausalLM.from_pretrained(model, quantization_config=bnb_config, device_map="cuda")
39
+
40
+ tokenizer = AutoTokenizer.from_pretrained(model, trust_remote_code=True)
41
+ tokenizer.pad_token = tokenizer.eos_token
42
+ tokenizer.padding_side = "right"
43
+
44
  prompt = f"""### Instruction:
45
  Below is an instruction that describes a task. Write a response that appropriately completes the request.
46