INDEX

Explanations

references to specific geographical locations or neighborhoods

oai_token-act-pair · gpt-3.5-turbo

mentions of neighborhoods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

inen

-1.08

indal

-0.83

uchin

-0.82

andum

-0.78

hide

-0.75

ulum

-0.75

emic

-0.74

stroke

-0.74

pas

-0.74

sonian

-0.73

POSITIVE LOGITS

 neighbourhood

1.05

 neighborhood

1.04

 neighborhoods

0.95

 neighbourhoods

0.90

 grocer

0.89

 Neigh

0.87

 Neighborhood

0.86

 streets

0.81

 neighbour

0.77

 neighbor

0.76

Activations Density 0.015%

references to specific geographical locations or neighborhoods

mentions of neighborhoods

No Comments

No Known Activations