INDEX

Explanations

mentions of the word "America" in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to "American" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

staking

-0.71

ĵĺ

-0.66

ĸļ

-0.65

 stocking

-0.63

pressed

-0.62

ŃĶ

-0.62

erest

-0.61

raw

-0.61

REE

-0.60

hem

-0.58

POSITIVE LOGITS

icans

1.31

ican

1.29

ICAN

1.25

ica

1.06

icas

0.97

sonian

0.97

icity

0.97

ilial

0.94

gency

0.88

gdala

0.88

Activations Density 0.008%

mentions of the word "America" in various contexts

references to "American" in various contexts

No Comments

No Known Activations