patulya commited on
Commit
c5cd957
1 Parent(s): 8610593

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +12 -13
README.md CHANGED
@@ -16,7 +16,7 @@ model = LLaMAForCausalLM.from_pretrained(
16
  "decapoda-research/llama-7b-hf",
17
  load_in_8bit=True,
18
  device_map="auto",
19
- )
20
  model = PeftModel.from_pretrained(model, "patulya/alpaca7B-lora")
21
 
22
  PROMPT = """Below is an instruction that describes a task. Write a response that appropriately completes the request.
@@ -31,21 +31,20 @@ inputs = tokenizer(
31
 
32
  input_ids = inputs["input_ids"].cuda()
33
 
34
- generation_config = GenerationConfig(
35
- temperature=0.6,
36
- top_p=0.95,
37
- repetition_penalty=1.15,
38
  )
39
 
40
  print("Generating...")
41
 
42
- generation_output = model.generate(
43
- input_ids=input_ids,
44
- generation_config=generation_config,
45
- return_dict_in_generate=True,
46
- output_scores=True,
47
- max_new_tokens=128,
48
  )
49
 
50
- for s in generation_output.sequences:
51
- print(tokenizer.decode(s))
 
16
  "decapoda-research/llama-7b-hf",
17
  load_in_8bit=True,
18
  device_map="auto",
19
+ )\
20
  model = PeftModel.from_pretrained(model, "patulya/alpaca7B-lora")
21
 
22
  PROMPT = """Below is an instruction that describes a task. Write a response that appropriately completes the request.
 
31
 
32
  input_ids = inputs["input_ids"].cuda()
33
 
34
+ generation_config = GenerationConfig(\
35
+ temperature=0.6,\
36
+ top_p=0.95,\
37
+ repetition_penalty=1.15,\
38
  )
39
 
40
  print("Generating...")
41
 
42
+ generation_output = model.generate(\
43
+ input_ids=input_ids,\
44
+ generation_config=generation_config,\
45
+ return_dict_in_generate=True,\
46
+ output_scores=True,\
47
+ max_new_tokens=128,\
48
  )
49
 
50
+ for s in generation_output.sequences: print(tokenizer.decode(s))