INDEX

Explanations

names of individuals

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

complete

-0.55

PDATE

-0.55

equality

-0.53

conference

-0.53

category

-0.53

piring

-0.53

REE

-0.53

 Newtown

-0.52

 Breaker

-0.52

RTX

-0.52

POSITIVE LOGITS

's

1.44

 himself

1.24

 herself

1.01

 Productions

0.99

ÃŃs

0.87

 remembers

0.86

Sr

0.85

Jr

0.85

 Enterprises

0.82

 Himself

0.81

Activations Density 0.232%

names of individuals

proper nouns, particularly names

No Comments

No Known Activations