INDEX

Explanations

mentions of the word "Bir" with varying activation values

oai_token-act-pair · gpt-3.5-turbo

words related to the term "biracial."

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

 nomine

-0.67

eur

-0.67

ĵĺ

-0.63

EMBER

-0.61

 Tune

-0.60

 paternal

-0.59

 tune

-0.58

 wise

-0.58

Unit

-0.58

 plat

-0.57

POSITIVE LOGITS

mingham

1.28

thing

1.08

git

1.05

ging

0.94

ney

0.92

chell

0.91

keley

0.91

combe

0.87

gel

0.87

ulia

0.86

Activations Density 0.021%

mentions of the word "Bir" with varying activation values

words related to the term "biracial."

No Comments

No Known Activations