INDEX

Explanations

facial expressions showing emotion, particularly focusing on eyes and faces

oai_token-act-pair · gpt-3.5-turbo

expressions of emotional reactions and facial responses

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

swick

-0.78

eor

-0.73

arro

-0.72

arthed

-0.72

ictionary

-0.71

tle

-0.70

ADA

-0.69

milo

-0.68

wald

-0.68

paces

-0.68

POSITIVE LOGITS

 demeanor

0.79

 onlook

0.71

animous

0.70

 palpable

0.69

 hearts

0.68

 disbelief

0.68

 faces

0.67

 purch

0.66

 emot

0.65

 neighbors

0.65

Activations Density 0.403%

facial expressions showing emotion, particularly focusing on eyes and faces

expressions of emotional reactions and facial responses

No Comments

No Known Activations