INDEX

Explanations

phrases related to specific actions or events that have occurred or will occur

oai_token-act-pair · gpt-3.5-turbo

repeated phrases indicating frequency or recurrence

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

CC

-0.67

aged

-0.64

 swat

-0.59

 cataly

-0.59

bud

-0.58

 steam

-0.58

 CPUs

-0.58

 drib

-0.58

 crush

-0.57

ding

-0.57

POSITIVE LOGITS

resa

1.15

romeda

1.08

odore

1.08

alyst

1.08

anmar

1.02

agascar

1.01

withstanding

0.96

ogether

0.94

icularly

0.93

pherd

0.92

Activations Density 0.698%

phrases related to specific actions or events that have occurred or will occur

repeated phrases indicating frequency or recurrence

No Comments

No Known Activations