INDEX

Explanations

sentences ending in '.'

oai_token-act-pair · gpt-3.5-turbo

phrases or references related to significant events or changes

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

 someday

-0.70

 discourage

-0.65

 pard

-0.64

 hope

-0.63

 welcomes

-0.63

morrow

-0.63

 urge

-0.62

 foreigners

-0.62

 nicer

-0.60

 wills

-0.60

POSITIVE LOGITS

âĢ

1.40

âĸ

1.18

Upon

1.02

Within

1.00

ãĢ

1.00

Ì¶

0.99

 âĢ

0.95

âĹ

0.94

Initially

0.92

ccording

0.92

Activations Density 0.628%

sentences ending in '.'

phrases or references related to significant events or changes

No Comments

No Known Activations