INDEX

Explanations

jersey-related terms and numbers within text

oai_token-act-pair · gpt-3.5-turbo

references to sports jerseys and their associated details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

yrinth

-0.90

lishes

-0.82

moderate

-0.78

enthal

-0.77

ricular

-0.75

audi

-0.74

raltar

-0.74

demon

-0.74

course

-0.74

ilib

-0.73

POSITIVE LOGITS

 jerseys

1.15

 jersey

1.01

seys

0.96

 worn

0.95

 tint

0.85

 sleeve

0.83

 sleeves

0.80

 sweater

0.80

 shirt

0.78

 shirts

0.78

Activations Density 0.013%

jersey-related terms and numbers within text

references to sports jerseys and their associated details

No Comments

No Known Activations