INDEX

Explanations

references to specific locations with the word "Grand."

oai_token-act-pair · gpt-3.5-turbo

references to "Grand" places or landmarks

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

Downloadha

-1.10

tics

-0.79

lihood

-0.77

PsyNetMessage

-0.76

ylon

-0.76

-0.70

utics

-0.70

-0.69

etimes

-0.68

 chemotherapy

-0.68

POSITIVE LOGITS

 Prix

1.18

father

1.12

mother

1.03

master

1.00

 Theft

0.98

 Rapids

0.98

parents

0.95

view

0.94

pa

0.90

iosity

0.87

Activations Density 0.009%

references to specific locations with the word "Grand."

references to "Grand" places or landmarks

No Comments

No Known Activations