INDEX

Explanations

names of people associated with criminal activities

oai_token-act-pair · gpt-3.5-turbo

mentions of criminal activities and related incidents

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Â£

-0.85

 Thatcher

-0.70

 scept

-0.67

 Britons

-0.66

 diarr

-0.63

 Â£

-0.63

 nerv

-0.62

fav

-0.62

 Whilst

-0.62

!:

-0.62

POSITIVE LOGITS

 Marketplace

0.73

 Lafayette

0.72

adena

0.71

iseum

0.71

 Monday

0.67

 Thursday

0.65

 Interstate

0.65

 Capitol

0.65

 Vance

0.65

ople

0.65

Activations Density 0.934%

names of people associated with criminal activities

mentions of criminal activities and related incidents

No Comments

No Known Activations