INDEX

Explanations

proper nouns

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Max" in various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

GROUND

-0.67

selves

-0.66

vag

-0.63

 referendum

-0.59

 revolving

-0.59

arers

-0.58

wark

-0.58

OPLE

-0.58

************

-0.57

 changing

-0.57

POSITIVE LOGITS

imus

1.52

imil

1.48

imize

1.42

imal

1.17

ima

1.13

ims

1.07

imo

1.06

ime

1.04

imum

1.00

imate

0.95

Activations Density 0.030%

proper nouns

mentions of the name "Max" in various contexts

No Comments

No Known Activations