INDEX

Explanations

phrases or sentences where the word "no" is prominently featured

oai_token-act-pair · gpt-3.5-turbo

instances of negation or refusal

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

Sensor

-0.63

ridor

-0.62

irez

-0.62

 Extras

-0.61

umen

-0.59

apeake

-0.59

andise

-0.59

arbon

-0.58

mud

-0.57

ortment

-0.57

POSITIVE LOGITS

 whatsoever

1.04

ody

0.81

urnal

0.69

onsense

0.68

nor

0.63

 dime

0.61

 hesitation

0.61

 coercion

0.60

 pudding

0.60

Ã¡s

0.59

Activations Density 0.199%

phrases or sentences where the word "no" is prominently featured

instances of negation or refusal

No Comments

No Known Activations