INDEX

Explanations

phrases related to a specific number or quantity

oai_token-act-pair · gpt-3.5-turbo

references to the word "one" 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 @ 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

ulsion

-0.65

older

-0.65

respective

-0.63

dk

-0.63

similar

-0.61

aeda

-0.60

FORMATION

-0.58

apons

-0.57

agall

-0.57

hess

-0.56

POSITIVE LOGITS

 sided

1.17

 dimensional

1.16

 hundred

0.96

0.93

 liner

0.93

dimensional

0.92

size

0.85

 ounce

0.84

 handed

0.83

âĢĳ

0.83

Activations Density 0.064%

phrases related to a specific number or quantity

references to the word "one" in various contexts

No Comments

No Known Activations