INDEX

Explanations

proper nouns related to a cryptocurrency called Vertcoin

oai_token-act-pair · gpt-3.5-turbo

references to a specific entity or brand involving "Vert" or similar terms

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

aeda

-0.82

usha

-0.69

riages

-0.67

ramid

-0.66

pper

-0.64

orters

-0.64

negie

-0.64

crew

-0.63

ourke

-0.62

 mash

-0.62

POSITIVE LOGITS

ible

1.10

ically

1.05

ibility

0.93

icals

0.93

worthiness

0.88

ibly

0.87

ured

0.83

ical

0.82

emis

0.80

ures

0.79

Activations Density 0.057%

proper nouns related to a cryptocurrency called Vertcoin

references to a specific entity or brand involving "Vert" or similar terms

No Comments

No Known Activations