INDEX

Explanations

phrases indicating a particular situation or context, such as "Amid" in news headlines

oai_token-act-pair · gpt-3.5-turbo

the word "Amid," especially in contexts related to chaos or controversy

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

 Mehran

-0.78

andering

-0.66

 detriment

-0.66

 station

-0.64

 hatch

-0.64

 manual

-0.61

 mosque

-0.61

ammers

-0.61

 checklist

-0.61

 obligation

-0.60

POSITIVE LOGITS

terness

0.98

ventory

0.85

ner

0.83

st

0.80

tha

0.77

²¾

0.77

Style

0.75

ĸļ

0.75

Discuss

0.75

 Amid

0.75

Activations Density 0.013%

phrases indicating a particular situation or context, such as "Amid" in news headlines

the word "Amid," especially in contexts related to chaos or controversy

No Comments

No Known Activations