INDEX

Explanations

indications about the direction or destination of actions or plans

oai_token-act-pair · gpt-3.5-turbo

expressions related to intentions and directions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

ALP

-0.73

=-=-

-0.67

 blat

-0.63

İĭ

-0.62

©¶æ

-0.62

 reinvest

-0.59

 Adin

-0.58

 paste

-0.58

 Hacker

-0.58

²

-0.57

POSITIVE LOGITS

/,

0.79

abouts

0.76

eter

0.75

igating

0.72

?:

0.71

!:

0.67

akeru

0.67

hammad

0.67

ilater

0.66

STER

0.66

Activations Density 0.133%

indications about the direction or destination of actions or plans

expressions related to intentions and directions

No Comments

No Known Activations