INDEX

Explanations

references to law enforcement activities and investigations

oai_token-act-pair · gpt-3.5-turbo

references to criminal investigations and suspects

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

 depl

-0.63

 Lust

-0.63

 Rape

-0.62

estern

-0.58

bleacher

-0.58

 Feminist

-0.57

,,,,,,,,

-0.56

ß

-0.56

 Brilliant

-0.55

 FANT

-0.55

POSITIVE LOGITS

 them

0.72

 afterward

0.71

 latter

0.71

 thereafter

0.69

 theirs

0.68

 otherwise

0.67

 specifics

0.67

 thereto

0.65

 later

0.64

 afterwards

0.63

Activations Density 1.331%

references to law enforcement activities and investigations

references to criminal investigations and suspects

No Comments

No Known Activations