INDEX

Explanations

names of individuals followed by the verb "wrote"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "wrote."

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

Ĭ±

-0.80

Enlarge

-0.73

phant

-0.66

nor

-0.66

aban

-0.66

aband

-0.65

abol

-0.64

ackle

-0.64

Lu

-0.63

abe

-0.63

POSITIVE LOGITS

 penned

0.90

smanship

0.85

 poems

0.84

writer

0.83

 written

0.82

wrote

0.81

 memos

0.78

 essays

0.78

 wrote

0.76

 aloud

0.76

Activations Density 0.033%

names of individuals followed by the verb "wrote"

instances of the word "wrote."

No Comments

No Known Activations