INDEX

Explanations

keywords related to the term "Wolverine"

oai_token-act-pair · gpt-3.5-turbo

references to the character Wolverine

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

avorite

-0.85

uct

-0.69

sembly

-0.68

uries

-0.66

classic

-0.65

holders

-0.65

 eclipse

-0.65

rophe

-0.64

elia

-0.64

wered

-0.63

POSITIVE LOGITS

OOL

0.96

 Wolver

0.93

sburg

0.92

inho

0.82

mingham

0.78

ines

0.71

kus

0.70

izont

0.70

ton

0.68

chy

0.67

Activations Density 0.019%

keywords related to the term "Wolverine"

references to the character Wolverine

No Comments

No Known Activations