INDEX

Explanations

mentions of the television network HBO

oai_token-act-pair · gpt-3.5-turbo

mentions of HBO and related content

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

acular

-0.86

uation

-0.82

acle

-0.80

omaly

-0.74

gew

-0.73

uart

-0.72

cking

-0.71

alion

-0.71

jury

-0.69

acea

-0.69

POSITIVE LOGITS

HBO

0.84

ÃįÃį

0.81

 Wire

0.81

 Films

0.80

TV

0.79

 premiered

0.78

 Networks

0.76

EW

0.76

IGHTS

0.74

BILL

0.72

Activations Density 0.007%

mentions of the television network HBO

mentions of HBO and related content

No Comments

No Known Activations