INDEX

Explanations

locations or landmarks

oai_token-act-pair · gpt-3.5-turbo

mentions of national parks and other national designations

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

icularly

-0.91

 Kinnikuman

-0.76

ablishment

-0.75

adish

-0.72

ppings

-0.71

oidal

-0.70

netflix

-0.68

ablished

-0.65

onal

-0.63

gotten

-0.63

POSITIVE LOGITS

 Speedway

0.82

 Resort

0.78

 Intermediate

0.70

search

0.69

aceutical

0.68

itent

0.68

 Parkway

0.68

astery

0.68

 Laboratory

0.67

 Prison

0.66

Activations Density 0.123%

locations or landmarks

mentions of national parks and other national designations

No Comments

No Known Activations