INDEX

Explanations

references to objects or actions related to wheels

oai_token-act-pair · gpt-3.5-turbo

references to wheels

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

ILA

-0.80

ISA

-0.77

 Sanford

-0.69

venants

-0.68

arant

-0.68

anta

-0.67

phia

-0.67

 Investigation

-0.66

Maria

-0.65

Kak

-0.64

POSITIVE LOGITS

 wheel

3.95

 Wheel

2.61

wheel

2.47

 wheels

2.34

Wheel

2.24

 Wheels

1.62

whe

1.42

 axle

1.40

 wheelchair

1.32

 pedal

1.28

Activations Density 0.009%

references to objects or actions related to wheels

references to wheels

No Comments

No Known Activations