INDEX

Explanations

references to a specific person with a name that includes "Ashraf" or "Varoufakis."

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of specific names and surnames

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Roma

-0.77

 Avalon

-0.75

 Parenthood

-0.69

AX

-0.66

 Genie

-0.65

 Stamford

-0.63

 Edison

-0.61

 Giovanni

-0.60

 Connor

-0.60

 Pompe

-0.60

POSITIVE LOGITS

shire

0.92

letcher

0.88

cffff

0.86

lisher

0.82

iture

0.82

idav

0.82

ugal

0.80

indal

0.80

cipled

0.76

entimes

0.75

Activations Density 0.036%

references to a specific person with a name that includes "Ashraf" or "Varoufakis."

repeated mentions of specific names and surnames

No Comments

No Known Activations