INDEX

Explanations

references to the word "Mon" followed by a single character

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "Mon" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Norn

-0.81

ACTED

-0.76

REE

-0.76

FUL

-0.72

ATING

-0.70

Reviewer

-0.69

IBLE

-0.69

ATOR

-0.69

ONSORED

-0.66

OHN

-0.66

POSITIVE LOGITS

itored

1.24

olith

1.19

stros

1.14

opoly

1.10

olithic

1.10

soon

1.10

astery

1.08

etary

1.01

roe

1.01

oton

1.00

Activations Density 0.016%

references to the word "Mon" followed by a single character

occurrences of the word "Mon" in various contexts

No Comments

No Known Activations