INDEX

Explanations

references to the city of Montreal

oai_token-act-pair · gpt-3.5-turbo

references to the city of Montreal

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

ullivan

-0.85

ramid

-0.84

================================================================

-0.77

rha

-0.75

omaly

-0.75

icio

-0.74

abet

-0.73

merce

-0.73

acea

-0.71

ordinate

-0.70

POSITIVE LOGITS

 Canadiens

1.15

 Gazette

1.07

 Neurolog

1.06

 Mirage

0.79

 Impact

0.77

 Protocol

0.77

ois

0.76

 Canad

0.75

 Montreal

0.74

 Mayor

0.74

Activations Density 0.011%

references to the city of Montreal

references to the city of Montreal

No Comments

No Known Activations