INDEX

Explanations

names of individuals in news articles

oai_token-act-pair · gpt-3.5-turbo

names of prominent individuals and their roles 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

ropolitan

-0.76

ournals

-0.75

ute

-0.74

ENDED

-0.69

enture

-0.69

ivities

-0.69

selves

-0.67

ired

-0.66

ENSE

-0.66

iences

-0.64

POSITIVE LOGITS

kus

0.99

 Ridley

0.98

rers

0.76

kj

0.71

lly

0.70

pillar

0.68

lings

0.68

 hatched

0.68

lett

0.68

 Ange

0.68

Activations Density 0.034%

names of individuals in news articles

names of prominent individuals and their roles in various contexts

No Comments

No Known Activations