INDEX

Explanations

words related to wheels and movement

oai_token-act-pair · gpt-3.5-turbo

references to wheels 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 @ 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

uates

-0.80

ropolitan

-0.79

nces

-0.77

uated

-0.77

ABE

-0.72

uation

-0.69

orescent

-0.69

uations

-0.67

alien

-0.67

stract

-0.67

POSITIVE LOGITS

chairs

1.34

chair

1.22

wheel

1.21

wright

1.20

wash

0.95

bar

0.90

Wheel

0.88

 wheel

0.87

base

0.86

house

0.84

Activations Density 0.015%

words related to wheels and movement

references to wheels in various contexts

No Comments

No Known Activations