INDEX

Explanations

references to the physical back or rear position of objects or people

oai_token-act-pair · gpt-3.5-turbo

references to the word "back"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.70

aukee

-0.70

Ń·

-0.69

 Parenthood

-0.69

inational

-0.68

izo

-0.67

upon

-0.66

Cola

-0.64

éĹĺ

-0.64

eteria

-0.64

POSITIVE LOGITS

side

1.18

packs

1.01

woods

1.00

drops

0.94

seat

0.90

 alley

0.89

hand

0.88

door

0.88

stab

0.88

gam

0.88

Activations Density 0.037%

references to the physical back or rear position of objects or people

references to the word "back"

No Comments

No Known Activations