INDEX

Explanations

phrases expressing identity or self-awareness

oai_token-act-pair · gpt-3.5-turbo

references to self-identity and personal experience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 retard

-0.66

arding

-0.58

CC

-0.56

uctions

-0.56

ongo

-0.56

 Voting

-0.55

 passing

-0.53

Edu

-0.53

 corrobor

-0.53

exec

-0.53

POSITIVE LOGITS

ufact

0.75

¥µ

0.67

shape

0.65

 Appears

0.64

areth

0.63

 selves

0.62

ues

0.61

ãĤ´ãĥ³

0.61

 entreprene

0.60

 Malk

0.60

Activations Density 0.253%

phrases expressing identity or self-awareness

references to self-identity and personal experience

No Comments

No Known Activations