INDEX

Explanations

names or words related to individuals such as "David" and "Hillary Clinton"

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names and identifiers related to individuals or entities

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

thora

-0.86

 exha

-0.77

 Rita

-0.75

vati

-0.73

 resil

-0.73

pmwiki

-0.72

psc

-0.71

 murd

-0.64

¹

-0.64

asma

-0.64

POSITIVE LOGITS

vironment

0.95

ovo

0.90

ception

0.85

heimer

0.83

iden

0.81

ciating

0.81

ly

0.80

sson

0.79

ally

0.77

structed

0.76

Activations Density 0.041%

names or words related to individuals such as "David" and "Hillary Clinton"

proper nouns, particularly names and identifiers related to individuals or entities

No Comments

No Known Activations