INDEX

Explanations

instances of the word "is" followed by a numerical value of 9 or 10

oai_token-act-pair · gpt-3.5-turbo

instances of the phrase "there is."

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

gypt

-0.72

buster

-0.65

 Mondays

-0.61

Hug

-0.60

bies

-0.58

wake

-0.57

 borough

-0.57

Think

-0.57

quote

-0.56

udeau

-0.55

POSITIVE LOGITS

 plenty

0.97

aido

0.85

å¥³

0.83

 ample

0.72

aternity

0.69

ample

0.69

 variability

0.68

 lots

0.66

ukong

0.66

peria

0.66

Activations Density 0.156%

instances of the word "is" followed by a numerical value of 9 or 10

instances of the phrase "there is."

No Comments

No Known Activations