INDEX

Explanations

phrases related to historical figures and scientific terms, potentially focusing on the suffix '-atus'

oai_token-act-pair · gpt-3.5-turbo

suffixes typically associated with scientific classifications or biological terms

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

sheet

-0.90

ishers

-0.75

¥ŀ

-0.72

ston

-0.68

street

-0.68

essert

-0.68

sheets

-0.66

ding

-0.66

ishing

-0.65

swe

-0.65

POSITIVE LOGITS

aurus

1.48

peria

1.02

CRIP

0.98

ource

0.97

xual

0.95

ocial

0.93

sein

0.90

xus

0.90

aeda

0.90

hift

0.89

Activations Density 0.037%

phrases related to historical figures and scientific terms, potentially focusing on the suffix '-atus'

suffixes typically associated with scientific classifications or biological terms

No Comments

No Known Activations