File size: 2,348 Bytes
8a20715
 
df6f372
 
 
 
 
 
 
 
8a20715
df6f372
 
 
05f1ff3
df6f372
 
 
 
 
3323075
df6f372
 
dbbc8bb
df6f372
dbbc8bb
 
3323075
df6f372
dbbc8bb
3323075
df6f372
3323075
 
 
 
313e8cc
dbbc8bb
4758b31
 
 
6735bb6
4758b31
 
 
 
 
 
dbbc8bb
78259e2
4758b31
 
 
 
 
 
 
 
 
 
 
dbbc8bb
 
79c0f84
 
 
 
cf3b11a
b142369
 
 
 
 
 
 
 
686c34c
b142369
686c34c
 
 
 
 
 
 
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
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
---
license: apache-2.0
language:
- en
tags:
- ablations
- datasets
- 1m
- llama
- custom_tokenizer
---

# DataBench

Rather than training one model, we trained multiple models, each on a different dataset. Everything else was kept the same, so that the only changing variable would be the dataset itself.

## Model Architecture

- Base Architecture: `LlamaForCasualLM`
- Tokenizer: [`Harley-ml/Dillionv2-1.3M`](https://huggingface.co/Harley-ml/Dillionv2-1.3M)
- Transformers Version: `5.13.1`
- Hidden Size: `128`
- Vocab Size: 2564
- Number of Layers: `8`
- Number of Heads: `4`
- Number of KV Heads: `2`
- Intermediate Size: `344`
- Head Dim: `32`
- Max Position Embeddings: `256`
- RoPe Theta: `2500.0`
- Tie Word Embeddings: `true`
- Hidden Activation: `silu`
- MLP Bias: `false`
- Initializer Range: `0.2`
- RMS Norm Eps: `1e-06`
- Pretraining Tp: `1`
- Use Cache: `false`
- **Total Parameters**: `1,780,352`

## Training Setup

- Epochs: `1`
- Max Steps: `-1.0`
- Batch Size: `400`
- Sequence Length: `256`
- Gradient Accumulation: `2`
- Gradient Clipping: `1.0`
- Gradient Checkpointing: `true`
- Learning Rate: `2.5e-3`
- Eval Split: `0.00165`
- Weight Decay: `0.01`
- Optimizer: `AdamW`
- AdamW Betas: `(0.9, 0.95)`
- AdamW Eps: `1e-8`
- Scheduler: `WSD`
- WSD Warmup Ratio: `0.015`
- WSD Stable Ratio: `0.78`
- WSD Decay Ratio: `0.20`
- WSD Minium LR Ratio: `0.0`
- WSD Number of Cycles: `0.5`
- DType: `float16`
- Torch.Compile: `true`
- DataLoader Workers: `2`
- Seed: `311`

# Results

Accuracy is normalized by length and shown as a percentage.
| Dataset           |   ARC-Easy |  HellaSwag |       PIQA |  **Avg ↑** |
| :---------------- | ---------: | ---------: | ---------: | ---------: |
| FineWeb           |     27.82% |     26.89% |     52.83% | **35.85%** |
| DCLM-1.0-Baseline |     29.08% |     27.06% |     52.23% | **36.12%** |
| DOAB              |     31.14% |     28.00% |     52.18% | **37.11%** |
| Project Gutenberg |     25.84% |     24.63% |     50.21% | **33.56%** |
| EOT-2004-Raw      |     29.08% |     27.54% |     51.31% | **35.98%** |
| Wikipedia         |     28.28% |     27.61% |     51.14% | **35.68%** |

(Will add ArithMark-3.0 later)
## Notice

This is a work in progress and is currently not completed. By the end of this project, we aim to have tested over 50 datasets.

## License

Apache 2.0.