INDEX

Explanations

instances of the word "how"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "how"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

isher

-0.66

ãĤ«

-0.60

idated

-0.60

yak

-0.58

Kou

-0.57

Et

-0.56

 Breaker

-0.56

agne

-0.55

-0.54

 Tanz

-0.54

POSITIVE LOGITS

soever

0.82

HCR

0.78

ever

0.74

ls

0.69

itzer

0.68

beit

0.68

 MUCH

0.65

ricanes

0.65

ihad

0.64

much

0.64

Activations Density 0.039%

instances of the word "how"

instances of the word "how"

No Comments

No Known Activations