INDEX

Explanations

legal or accusatory language

oai_token-act-pair · gpt-3.5-turbo

phrases related to demands and actions taken for accountability or justice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

amorph

-0.64

riched

-0.63

 blending

-0.61

omics

-0.60

 superpower

-0.60

throp

-0.60

ennial

-0.59

ominated

-0.58

pires

-0.58

Enlarge

-0.57

POSITIVE LOGITS

 DMCA

0.91

 apology

0.89

 refund

0.87

 apologised

0.84

 apologise

0.84

 refunds

0.83

 admins

0.81

 apologies

0.78

 whistlebl

0.76

 lawy

0.75

Activations Density 1.718%

legal or accusatory language

phrases related to demands and actions taken for accountability or justice

No Comments

No Known Activations