INDEX

Explanations

mentions of specific locations or places, specifically parking lots

oai_token-act-pair · gpt-3.5-turbo

references to parking lots

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

sonian

-0.90

irtual

-0.71

utenberg

-0.71

 fictional

-0.68

Narr

-0.68

nom

-0.67

untu

-0.66

anish

-0.66

 microscopic

-0.65

umar

-0.65

POSITIVE LOGITS

stones

0.79

lot

0.78

tery

0.77

Lot

0.76

Lot

0.74

ages

0.74

esa

0.73

Nig

0.72

 attendants

0.72

aurus

0.68

Activations Density 0.017%

mentions of specific locations or places, specifically parking lots

references to parking lots

No Comments

No Known Activations