INDEX

Explanations

names of individuals or entities ending with "n"

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter 'n' in the text

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

xual

-0.78

 welf

-0.66

 caution

-0.64

 compe

-0.64

EStreamFrame

-0.60

 mosqu

-0.60

 DRAG

-0.60

 pissed

-0.59

 hypot

-0.59

 conduc

-0.59

POSITIVE LOGITS

uggets

1.17

ihil

1.13

aturally

1.11

onsense

1.11

ovation

1.08

ucle

1.06

umerous

1.05

orthern

1.01

acht

1.01

swer

0.99

Activations Density 0.045%

names of individuals or entities ending with "n"

occurrences of the letter 'n' in the text

No Comments

No Known Activations