INDEX

Explanations

mentions of a specific location, likely related to news or events happening in that area

oai_token-act-pair · gpt-3.5-turbo

mentions of specific place names or locations

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

Magikarp

-0.92

natureconservancy

-0.76

ittal

-0.72

ulic

-0.70

mediated

-0.69

clerosis

-0.68

externalActionCode

-0.67

uana

-0.66

utic

-0.66

ptive

-0.65

POSITIVE LOGITS

ington

0.97

 Hills

0.81

mouth

0.79

dale

0.77

 Cros

0.77

igans

0.76

INGTON

0.74

 Avenue

0.74

stown

0.73

 Coliseum

0.73

Activations Density 0.021%

mentions of a specific location, likely related to news or events happening in that area

mentions of specific place names or locations

No Comments

No Known Activations