INDEX

Explanations

phrases containing the term "aus" with varying levels of intensity

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of "aus" or "source."

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

 cane

-0.63

 receptor

-0.62

 observable

-0.62

 short

-0.61

 master

-0.61

oted

-0.59

 CARD

-0.59

 average

-0.59

 yield

-0.59

dfx

-0.59

POSITIVE LOGITS

aus

1.55

chwitz

1.06

uin

1.01

atsu

1.00

ername

0.99

ertodd

0.97

hent

0.95

edIn

0.94

nown

0.94

uve

0.93

Activations Density 0.006%

phrases containing the term "aus" with varying levels of intensity

words related to the concept of "aus" or "source."

No Comments

No Known Activations