INDEX

Explanations

mentions of the word "orn"

oai_token-act-pair · gpt-3.5-turbo

mentions of unicorns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

dule

-0.76

Pwr

-0.73

perature

-0.71

 refreshing

-0.67

 Lowry

-0.61

MIA

-0.61

pton

-0.60

 Quincy

-0.60

 honestly

-0.60

arnaev

-0.60

POSITIVE LOGITS

orn

1.02

obyl

0.96

odon

0.95

alia

0.87

quist

0.78

ication

0.77

ail

0.76

usa

0.76

OGR

0.75

ography

0.75

Activations Density 0.008%

mentions of the word "orn"

mentions of unicorns

No Comments

No Known Activations