Datasets:
pingjiewang commited on
Commit ·
ffed732
1
Parent(s): 34fa27c
Rename to stepwise-token-learning: update title, links and citation
Browse files- CITATION.cff +2 -2
- README.md +4 -4
- manifest.yaml +2 -2
CITATION.cff
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
cff-version: 1.2.0
|
| 2 |
message: "If you use these loss matrices, please cite the accompanying paper."
|
| 3 |
-
title: "Token Learning
|
| 4 |
type: dataset
|
| 5 |
abstract: "Per-token validation-loss trajectories for one million contextualized tokens along the training-step, data-scale, and model-scale axes of LLM pre-training, plus a synthetic arithmetic example."
|
| 6 |
keywords:
|
|
@@ -20,7 +20,7 @@ authors:
|
|
| 20 |
- family-names: Zhang
|
| 21 |
given-names: Debing
|
| 22 |
license: MIT
|
| 23 |
-
url: "https://huggingface.co/datasets/applewpj/token-learning
|
| 24 |
repository-code: "https://github.com/applewpj/stepwise-token-learning"
|
| 25 |
preferred-citation:
|
| 26 |
type: conference-paper
|
|
|
|
| 1 |
cff-version: 1.2.0
|
| 2 |
message: "If you use these loss matrices, please cite the accompanying paper."
|
| 3 |
+
title: "Stepwise Token Learning: Token-Level Loss Trajectories Along Scaling Axes"
|
| 4 |
type: dataset
|
| 5 |
abstract: "Per-token validation-loss trajectories for one million contextualized tokens along the training-step, data-scale, and model-scale axes of LLM pre-training, plus a synthetic arithmetic example."
|
| 6 |
keywords:
|
|
|
|
| 20 |
- family-names: Zhang
|
| 21 |
given-names: Debing
|
| 22 |
license: MIT
|
| 23 |
+
url: "https://huggingface.co/datasets/applewpj/stepwise-token-learning"
|
| 24 |
repository-code: "https://github.com/applewpj/stepwise-token-learning"
|
| 25 |
preferred-citation:
|
| 26 |
type: conference-paper
|
README.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
-
pretty_name: "Token Learning
|
| 4 |
task_categories:
|
| 5 |
- text-generation
|
| 6 |
tags:
|
|
@@ -50,7 +50,7 @@ configs:
|
|
| 50 |
<a href="#citation"><img src="https://raw.githubusercontent.com/applewpj/stepwise-token-learning/main/assets/readme/btn-bibtex.svg" alt="BibTeX"></a>
|
| 51 |
</p>
|
| 52 |
|
| 53 |
-
# Token Learning
|
| 54 |
|
| 55 |
Per-token validation-loss trajectories for **one million contextualized tokens**
|
| 56 |
along each of three scaling axes of LLM pre-training (training steps *T*, data
|
|
@@ -109,7 +109,7 @@ Browse a preview with 🤗 Datasets:
|
|
| 109 |
```python
|
| 110 |
from datasets import load_dataset
|
| 111 |
|
| 112 |
-
repo = "applewpj/token-learning
|
| 113 |
preview = load_dataset(repo, "step_axis_1m", split="preview") # columns: row, loss[54]
|
| 114 |
axes = load_dataset(repo, "axes", split="full") # axis values per asset
|
| 115 |
```
|
|
@@ -121,7 +121,7 @@ import numpy as np
|
|
| 121 |
from huggingface_hub import hf_hub_download
|
| 122 |
|
| 123 |
path = hf_hub_download(
|
| 124 |
-
repo_id="applewpj/token-learning
|
| 125 |
repo_type="dataset",
|
| 126 |
filename="examples/step_axis_1m/losses.npz",
|
| 127 |
)
|
|
|
|
| 1 |
---
|
| 2 |
license: mit
|
| 3 |
+
pretty_name: "Stepwise Token Learning: Token-Level Loss Trajectories Along Scaling Axes"
|
| 4 |
task_categories:
|
| 5 |
- text-generation
|
| 6 |
tags:
|
|
|
|
| 50 |
<a href="#citation"><img src="https://raw.githubusercontent.com/applewpj/stepwise-token-learning/main/assets/readme/btn-bibtex.svg" alt="BibTeX"></a>
|
| 51 |
</p>
|
| 52 |
|
| 53 |
+
# Stepwise Token Learning: Token-Level Loss Trajectories
|
| 54 |
|
| 55 |
Per-token validation-loss trajectories for **one million contextualized tokens**
|
| 56 |
along each of three scaling axes of LLM pre-training (training steps *T*, data
|
|
|
|
| 109 |
```python
|
| 110 |
from datasets import load_dataset
|
| 111 |
|
| 112 |
+
repo = "applewpj/stepwise-token-learning"
|
| 113 |
preview = load_dataset(repo, "step_axis_1m", split="preview") # columns: row, loss[54]
|
| 114 |
axes = load_dataset(repo, "axes", split="full") # axis values per asset
|
| 115 |
```
|
|
|
|
| 121 |
from huggingface_hub import hf_hub_download
|
| 122 |
|
| 123 |
path = hf_hub_download(
|
| 124 |
+
repo_id="applewpj/stepwise-token-learning",
|
| 125 |
repo_type="dataset",
|
| 126 |
filename="examples/step_axis_1m/losses.npz",
|
| 127 |
)
|
manifest.yaml
CHANGED
|
@@ -1,7 +1,7 @@
|
|
| 1 |
version: 1
|
| 2 |
repo_type: dataset
|
| 3 |
-
recommended_repo_name: token-learning
|
| 4 |
-
description: Public loss matrices for reproducing Token Learning
|
| 5 |
|
| 6 |
files:
|
| 7 |
- id: step_axis_1m
|
|
|
|
| 1 |
version: 1
|
| 2 |
repo_type: dataset
|
| 3 |
+
recommended_repo_name: stepwise-token-learning
|
| 4 |
+
description: Public loss matrices for reproducing the Stepwise Token Learning figures.
|
| 5 |
|
| 6 |
files:
|
| 7 |
- id: step_axis_1m
|