INDEX

Explanations

statements or quotes spoken by someone

oai_token-act-pair · gpt-3.5-turbo

instances of dialogue or statements made by individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ãĥİ

-0.72

thur

-0.65

ãĥĥãĥĪ

-0.65

paralle

-0.63

MU

-0.62

arent

-0.62

 pend

-0.60

å§«

-0.59

 ILCS

-0.59

resent

-0.59

POSITIVE LOGITS

 sarcast

1.06

 bluntly

1.02

 rhet

0.96

 emphatically

0.89

 referring

0.83

 afterward

0.81

 diplom

0.78

 aloud

0.78

 incred

0.77

 proudly

0.75

Activations Density 0.123%

statements or quotes spoken by someone

instances of dialogue or statements made by individuals

No Comments

No Known Activations