INDEX

Explanations

words related to disappearance or absence

oai_token-act-pair · gpt-3.5-turbo

the occurrence of the word "gone."

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

orsi

-0.78

ullah

-0.77

ounters

-0.75

ussen

-0.75

eers

-0.72

enegger

-0.71

chev

-0.70

ellation

-0.70

itudes

-0.69

arya

-0.69

POSITIVE LOGITS

 Away

0.89

 unnoticed

0.79

 Gone

0.79

 overboard

0.79

AW

0.76

 away

0.75

 viral

0.73

Bye

0.73

MAD

0.73

HAM

0.72

Activations Density 0.019%

words related to disappearance or absence

the occurrence of the word "gone."

No Comments

No Known Activations