INDEX

Explanations

proper nouns related to locations or institutions

oai_token-act-pair · gpt-3.5-turbo

the special character tokens indicating the end of the document and references to "Saint" followed by a number

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ical

-0.70

sequence

-0.66

iferation

-0.63

clips

-0.63

HO

-0.63

mable

-0.60

itsch

-0.60

idation

-0.59

 sled

-0.58

PT

-0.58

POSITIVE LOGITS

 Petersburg

1.04

 Laurent

1.02

clair

1.00

 Clair

0.93

 Augustine

0.89

 Louis

0.89

 Lucia

0.88

 Kitt

0.85

 Francis

0.82

onew

0.81

Activations Density 0.025%

proper nouns related to locations or institutions

the special character tokens indicating the end of the document and references to "Saint" followed by a number

No Comments

No Known Activations