INDEX

Explanations

locations or establishments with the word 'diner'

oai_token-act-pair · gpt-3.5-turbo

variations of the word "diner"

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

hovah

-0.96

ecause

-0.71

isSpecialOrderable

-0.71

ORK

-0.70

gap

-0.70

lished

-0.69

soDeliveryDate

-0.67

nep

-0.67

akening

-0.66

rouse

-0.66

POSITIVE LOGITS

iner

1.19

iners

0.91

ator

0.81

ators

0.79

TAIN

0.79

throp

0.77

ogly

0.76

ception

0.74

 offic

0.74

aton

0.74

Activations Density 0.014%

locations or establishments with the word 'diner'

variations of the word "diner"

No Comments

No Known Activations