INDEX

Explanations

mentions of the pronoun "Us" with varying levels of relevance

oai_token-act-pair · gpt-3.5-turbo

references to "Us" 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 @ 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

served

-0.76

hetti

-0.63

CPC

-0.62

 confinement

-0.62

 lure

-0.61

 Clarkson

-0.61

 chaired

-0.60

*/(

-0.59

 runoff

-0.56

 lull

-0.56

POSITIVE LOGITS

hers

1.20

agi

1.09

ages

0.93

urers

0.92

ern

0.91

uman

0.91

chwitz

0.90

selves

0.88

ername

0.87

umi

0.86

Activations Density 0.019%

mentions of the pronoun "Us" with varying levels of relevance

references to "Us" in various contexts

No Comments

No Known Activations