INDEX

Explanations

indicators of news articles and their sources

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Beam

-0.70

 Astral

-0.70

GSL

-0.68

Sly

-0.64

Xan

-0.62

 Haku

-0.61

 Serve

-0.60

 Curse

-0.60

 Clim

-0.60

 Pyramid

-0.60

POSITIVE LOGITS

news

1.16

politics

1.09

articles

1.07

article

1.04

1.03

features

0.99

column

0.99

stories

0.99

0.98

middle

0.98

Activations Density 0.025%

indicators of news articles and their sources

No Comments

No Known Activations