INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns and specific names related to individuals or specific entities

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

berman

-0.77

 Fathers

-0.67

onut

-0.66

omez

-0.63

 dating

-0.61

 pseudonym

-0.59

İĭ

-0.59

opez

-0.59

emort

-0.58

cab

-0.58

POSITIVE LOGITS

IELD

0.94

âķĲâķĲ

0.87

IFT

0.85

ments

0.84

adow

0.82

ings

0.82

resh

0.81

itic

0.77

adows

0.75

aun

0.73

Activations Density 0.008%

names of individuals

proper nouns and specific names related to individuals or specific entities

No Comments

No Known Activations