File size: 1,626 Bytes
e5f699c
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
4a9f785
 
 
 
0b4ae8c
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
---
dataset_info:
  features:
  - name: is_thoughtful
    dtype: bool
  - name: uuid
    dtype: string
  - name: chunk_index
    dtype: int64
  - name: chunk_start_index
    dtype: int64
  - name: chunk_end_index
    dtype: int64
  - name: num_scored_tokens
    dtype: int64
  - name: input_ids
    list:
      list: int32
  - name: thought_text
    list: large_string
  - name: g
    list: int64
  - name: truncated
    list: bool
  - name: thoughtless_log_prob_sum
    dtype: float32
  - name: thoughtless_log_prob_per_token
    list: float32
  - name: thoughtful_log_prob_sum
    list: float32
  - name: thoughtful_log_prob_per_token
    list:
      list: float32
  splits:
  - name: train
    num_bytes: 93960828.5
    num_examples: 311296
  - name: test
    num_bytes: 93960828.5
    num_examples: 311296
  download_size: 69936144
  dataset_size: 187921657.0
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: test
    path: data/test-*
---
Example of a chunk record dataset, made for debugging purposes.

Made by taking the step-1216 validation chunk records of [reason-only-k=32-lr=0.00015-replay=0.5](https://wandb.ai/latent-thoughts/prestar/runs/rb7kxdbd?nw=nwuserjackellishsieh) and splitting the doc uuids in half into train and test. Thus, all of a document's chunks lie within exactly one of the splits.

There are (4096/32) * 4864 =622\_592$ chunks (including the first chunk) collectively.

| Split | Chunks|Thoughtful Chunks|Thoughtful Proportion
|-|-|-|-|
|`train`|311_296|38_557 | 0.1239 |
|`test`|311_296| 38_659 | 0.1242 |
| Total | 622_592 | 77_216 | 0.1240 |