pszemraj commited on
Commit
cb196c1
·
1 Parent(s): 82f65e2

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +8 -4
README.md CHANGED
@@ -7,21 +7,25 @@ metrics:
7
  model-index:
8
  - name: BL-pythia-31m-simple_wikipedia_LM-2048-scratch
9
  results: []
 
 
 
 
10
  ---
11
 
12
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
  should probably proofread and complete it, then remove this comment. -->
14
 
15
- # BL-pythia-31m-simple_wikipedia_LM-2048-scratch
16
 
17
- This model is a fine-tuned version of [EleutherAI/pythia-31m](https://huggingface.co/EleutherAI/pythia-31m) on the None dataset.
18
  It achieves the following results on the evaluation set:
19
  - Loss: 4.1763
20
  - Accuracy: 0.3676
21
 
22
  ## Model description
23
 
24
- More information needed
25
 
26
  ## Intended uses & limitations
27
 
@@ -71,4 +75,4 @@ The following hyperparameters were used during training:
71
  - Transformers 4.33.1
72
  - Pytorch 2.2.0.dev20230907+cu118
73
  - Datasets 2.14.5
74
- - Tokenizers 0.13.3
 
7
  model-index:
8
  - name: BL-pythia-31m-simple_wikipedia_LM-2048-scratch
9
  results: []
10
+ license: apache-2.0
11
+ datasets:
12
+ - pszemraj/simple_wikipedia_LM
13
+ pipeline_tag: text-generation
14
  ---
15
 
16
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
17
  should probably proofread and complete it, then remove this comment. -->
18
 
19
+ # pythia-31m-simplewiki-scratch-bf16
20
 
21
+ Trained from random initialized config based on [EleutherAI/pythia-31m](https://huggingface.co/EleutherAI/pythia-31m), 3 epochs bf16
22
  It achieves the following results on the evaluation set:
23
  - Loss: 4.1763
24
  - Accuracy: 0.3676
25
 
26
  ## Model description
27
 
28
+ tuned with bf16 (previous was fp32)
29
 
30
  ## Intended uses & limitations
31
 
 
75
  - Transformers 4.33.1
76
  - Pytorch 2.2.0.dev20230907+cu118
77
  - Datasets 2.14.5
78
+ - Tokenizers 0.13.3