INDEX

Explanations

the word "The."

oai_token-act-pair · gpt-3.5-turbo

the recurring phrase "The" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ement

-0.69

poke

-0.68

pecially

-0.62

 beforehand

-0.60

gpu

-0.59

thood

-0.59

actory

-0.58

 entertained

-0.58

 until

-0.58

omever

-0.58

POSITIVE LOGITS

oret

1.40

resa

1.17

ories

1.07

odore

1.04

atre

1.01

sis

0.98

orem

0.97

 Basics

0.96

 Economist

0.96

odor

0.94

Activations Density 0.350%

the word "The."

the recurring phrase "The" in various contexts

No Comments

No Known Activations