INDEX

Explanations

proper nouns preceded by 'who'

oai_token-act-pair · gpt-3.5-turbo

instances of the word "who."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

AIR

-0.73

ulp

-0.68

AN

-0.68

apest

-0.66

hazard

-0.66

Sharp

-0.65

CONCLUS

-0.65

uta

-0.64

"(

-0.63

PORT

-0.63

POSITIVE LOGITS

 redes

0.86

ophon

0.76

 frequ

0.75

 accompanies

0.75

 spearheaded

0.74

 oversaw

0.73

 oversees

0.73

 owns

0.72

 grew

0.71

 operates

0.70

Activations Density 0.102%

proper nouns preceded by 'who'

instances of the word "who."

No Comments

No Known Activations