INDEX

Explanations

proper nouns related to a specific person or location

oai_token-act-pair · gpt-3.5-turbo

occurrences of a specific term or phrase related to a category, possibly related to "ga" as a theme across various 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lessly

-0.83

TY

-0.72

tle

-0.71

paren

-0.71

CAST

-0.70

landish

-0.70

neys

-0.69

andem

-0.68

Topics

-0.68

ty

-0.66

POSITIVE LOGITS

vernment

1.03

veyard

0.94

ussian

0.83

Ã±a

0.82

arb

0.81

Ga

0.81

arde

0.80

ffiti

0.80

uthor

0.76

xon

0.76

Activations Density 0.010%

proper nouns related to a specific person or location

occurrences of a specific term or phrase related to a category, possibly related to "ga" as a theme across various contexts

No Comments

No Known Activations