INDEX

Explanations

references to World War II (WWII) and related historical or military terms

oai_token-act-pair · gpt-3.5-turbo

mentions of the acronym "WW" and its related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Downloadha

-0.84

 succeeding

-0.80

uated

-0.76

alties

-0.73

staking

-0.71

Reviewer

-0.70

uate

-0.69

ioned

-0.66

 yielding

-0.66

etsk

-0.66

POSITIVE LOGITS

WW

1.26

ombat

0.94

WF

0.85

JD

0.85

icz

0.81

restling

0.81

WW

0.80

AMP

0.80

urst

0.79

 Norton

0.79

Activations Density 0.004%

references to World War II (WWII) and related historical or military terms

mentions of the acronym "WW" and its related contexts

No Comments

No Known Activations