INDEX

Explanations

specific locations and place names, particularly in Ontario, Canada

oai_token-act-pair · gpt-3.5-turbo

geographic locations, specifically names of places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ebus

-1.01

Reviewer

-0.89

ources

-0.82

Las

-0.81

ãĤ·ãĥ£

-0.79

VD

-0.77

áµ

-0.74

Pand

-0.73

aceae

-0.70

NRS

-0.70

POSITIVE LOGITS

shire

0.98

 harbour

0.90

 neighbourhoods

0.88

onge

0.88

auga

0.87

 Borough

0.83

 Centre

0.80

 neighbourhood

0.79

 Mississ

0.78

 Heights

0.78

Activations Density 0.031%

specific locations and place names, particularly in Ontario, Canada

geographic locations, specifically names of places

No Comments

No Known Activations