INDEX

Explanations

mentions of Twitter activity

oai_token-act-pair · gpt-3.5-turbo

references to social media activity, especially on Twitter

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.10.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 asbestos

-0.88

iHUD

-0.80

 appendix

-0.77

CRC

-0.76

pdf

-0.75

bestos

-0.72

 Auditor

-0.71

 Siem

-0.69

 Bagg

-0.68

 examinations

-0.68

POSITIVE LOGITS

 retweet

2.32

 tweets

2.24

 hasht

2.12

 tweet

2.07

 tweeting

2.05

 Twitter

2.00

 hashtag

1.95

Twitter

1.88

 twitter

1.86

 tweeted

1.85

Activations Density 0.445%

mentions of Twitter activity

references to social media activity, especially on Twitter

No Comments

No Known Activations