INDEX

Explanations

phrases related to processes or actions

oai_token-act-pair · gpt-3.5-turbo

phrases related to processes and their outcomes

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

 leaflets

-0.69

 needle

-0.57

cial

-0.57

 Frequ

-0.56

 strongly

-0.55

 Firearms

-0.55

 copyrighted

-0.54

 variable

-0.54

 flashlight

-0.53

 finger

-0.53

POSITIVE LOGITS

 thereof

1.12

ãĤ©

0.93

ogether

0.80

vre

0.78

APTER

0.77

 therein

0.76

OTAL

0.75

result

0.75

 attRot

0.74

 meantime

0.74

Activations Density 0.711%

phrases related to processes or actions

phrases related to processes and their outcomes

No Comments

No Known Activations