INDEX

Explanations

dates specified with the day of the week for public figures

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "on" 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

ENCY

-0.66

rador

-0.65

hetically

-0.65

APH

-0.60

oother

-0.60

ACTED

-0.60

RIC

-0.59

posed

-0.58

auga

-0.58

acter

-0.58

POSITIVE LOGITS

Tue

1.05

Feb

1.05

Apr

1.03

Aug

0.99

Thu

0.96

Jul

0.95

 behalf

0.94

Nov

0.94

Sep

0.92

 March

0.90

Activations Density 0.037%

dates specified with the day of the week for public figures

occurrences of the word "on" in various contexts

No Comments

No Known Activations