INDEX

Explanations

words related to a specific location or entity named "Wake."

oai_token-act-pair · gpt-3.5-turbo

occurrences of the term "Wake" and variations or phrases related to it

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

bably

-0.73

served

-0.70

igree

-0.68

ioned

-0.67

rador

-0.66

afort

-0.66

inates

-0.64

 fractures

-0.62

atem

-0.62

displayText

-0.62

POSITIVE LOGITS

 Wake

1.36

fulness

0.90

wake

0.89

mares

0.85

gate

0.84

caster

0.81

urst

0.78

wash

0.78

 Chal

0.77

Til

0.73

Activations Density 0.004%

words related to a specific location or entity named "Wake."

occurrences of the term "Wake" and variations or phrases related to it

No Comments

No Known Activations