INDEX

Explanations

names and locations, particularly mentioning an individual or place names in a statement

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of people and places

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

acted

-0.69

 stakes

-0.65

growth

-0.64

lest

-0.64

 earth

-0.64

DERR

-0.63

 suspense

-0.60

acts

-0.60

 plur

-0.59

Pg

-0.57

POSITIVE LOGITS

hof

0.80

wagen

0.80

heid

0.75

abee

0.71

imore

0.70

erton

0.70

ascus

0.69

illac

0.68

ayn

0.68

ulhu

0.66

Activations Density 0.258%

names and locations, particularly mentioning an individual or place names in a statement

proper nouns, particularly names of people and places

No Comments

No Known Activations