INDEX

Explanations

mentions of neighboring individuals or actions related to being a neighbor

oai_token-act-pair · gpt-3.5-turbo

references to neighbors

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

inen

-0.88

anche

-0.85

othal

-0.79

nesota

-0.78

oola

-0.76

otine

-0.76

uggage

-0.75

ulum

-0.75

indal

-0.72

emia

-0.72

POSITIVE LOGITS

 neighbor

1.12

 neighbors

1.11

 Neigh

1.08

 neighbours

1.04

 neighbour

1.04

 Neighbor

0.87

bors

0.83

liness

0.80

Neigh

0.79

folk

0.77

Activations Density 0.018%

mentions of neighboring individuals or actions related to being a neighbor

references to neighbors

No Comments

No Known Activations