INDEX

Explanations

phrases related to explaining or justifying something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "explain" and its variants

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

luster

-0.74

assic

-0.74

itton

-0.69

rica

-0.65

sembly

-0.65

-0.64

quart

-0.64

rection

-0.63

Pont

-0.63

isol

-0.63

POSITIVE LOGITS

why

1.79

WHY

1.46

how

1.39

why

1.35

Why

1.00

 away

1.00

how

1.00

HOW

0.99

 what

0.96

 whence

0.92

Activations Density 0.075%

phrases related to explaining or justifying something

instances of the word "explain" and its variants

No Comments

No Known Activations