INDEX

Explanations

locations or places

oai_token-act-pair · gpt-3.5-turbo

proper nouns, specifically names of places and organizations

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

umbnails

-0.83

apego

-0.79

enegger

-0.76

icators

-0.72

rogens

-0.71

ographers

-0.67

vous

-0.64

 bearer

-0.64

quished

-0.62

ournal

-0.61

POSITIVE LOGITS

 guise

0.88

 vicinity

0.87

town

0.84

 Heights

0.84

 Meadow

0.81

 Meadows

0.79

dale

0.79

 contexts

0.78

 Alley

0.77

cember

0.76

Activations Density 0.307%

locations or places

proper nouns, specifically names of places and organizations

No Comments

No Known Activations