alonzogarbanzo commited on
Commit
dfec861
1 Parent(s): 01995b2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +21 -3
README.md CHANGED
@@ -22,12 +22,30 @@ Intended for use on a student group project for Portland State University's Wint
22
 
23
  ## Training and evaluation data
24
 
25
- More information needed
26
 
27
  ## Training procedure
28
 
29
  Trained on a single RTX 3090 card.
30
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
31
  ### Training hyperparameters
32
 
33
  The following hyperparameters were used during training:
@@ -43,10 +61,10 @@ The following hyperparameters were used during training:
43
  - mixed_precision_training: Native AMP
44
 
45
  ### Training results
46
- After final epoch:
47
  {'loss': 0.0472, 'learning_rate': 1.4893617021276598e-06, 'epoch': 4.95}
48
 
49
- After full completion
50
  {'train_runtime': 563.2707,
51
  'train_samples_per_second': 1.687,
52
  'train_steps_per_second': 0.417,
 
22
 
23
  ## Training and evaluation data
24
 
25
+ Instruction Tuned on the creative writing dataset here: https://huggingface.co/datasets/adambjorn/UnrelatedForgettingOverhead/viewer/creative
26
 
27
  ## Training procedure
28
 
29
  Trained on a single RTX 3090 card.
30
 
31
+ Given a set of prompts:
32
+
33
+ ```python
34
+ prompts = [
35
+ "Write a creative short story based on the following title:",
36
+ "Here is a title for a story. Craft a short narrative around it:",
37
+ "Using the title given, develop a short story:",
38
+ "Imagine a short story that starts with this title:",
39
+ "Create a brief story with the following title:"
40
+ ]
41
+ ```
42
+
43
+ Concatenate the prompt, the title and the story like so:
44
+
45
+ ```python
46
+ concatenated_texts = [random.choice(prompts) + " " + title + "</s>" + "Story: " + selftext for title, selftext in zip(titles, selftexts)]
47
+ ```
48
+
49
  ### Training hyperparameters
50
 
51
  The following hyperparameters were used during training:
 
61
  - mixed_precision_training: Native AMP
62
 
63
  ### Training results
64
+ Final results:
65
  {'loss': 0.0472, 'learning_rate': 1.4893617021276598e-06, 'epoch': 4.95}
66
 
67
+ Average results:
68
  {'train_runtime': 563.2707,
69
  'train_samples_per_second': 1.687,
70
  'train_steps_per_second': 0.417,