INDEX

Explanations

mentions of the word "Lombardi"

oai_token-act-pair · gpt-3.5-turbo

mentions of a specific individual named Lomb

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

xual

-0.96

terday

-0.80

=~=~

-0.77

SEA

-0.75

earcher

-0.74

ãģĦ

-0.69

pecting

-0.69

ctory

-0.69

rgb

-0.68

DOWN

-0.67

POSITIVE LOGITS

ories

0.79

oise

0.78

orem

0.78

atures

0.72

ature

0.72

amia

0.72

atum

0.71

ively

0.71

ardy

0.70

icity

0.70

Activations Density 0.035%

mentions of the word "Lombardi"

mentions of a specific individual named Lomb

No Comments

No Known Activations