INDEX

Explanations

phrases related to decision-making

oai_token-act-pair · gpt-3.5-turbo

statements about subjective opinions or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

mares

-0.86

ciating

-0.74

-0.73

ãĤ©

-0.72

details

-0.71

azon

-0.69

igs

-0.69

arthed

-0.68

igg

-0.65

ffe

-0.64

POSITIVE LOGITS

 appropriate

1.49

 prudent

1.44

 advisable

1.42

 worthwhile

1.39

 necessary

1.34

 preferable

1.31

 desirable

1.28

appropriate

1.26

 acceptable

1.25

 advantageous

1.25

Activations Density 0.318%

phrases related to decision-making

statements about subjective opinions or evaluations

No Comments

No Known Activations