INDEX

Explanations

references to smiles

oai_token-act-pair · gpt-3.5-turbo

references to smiles and expressions of happiness

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

æ©Ł

-0.79

raped

-0.73

 Administ

-0.72

FER

-0.71

vernight

-0.70

Ranked

-0.69

aer

-0.69

needs

-0.67

assic

-0.65

lay

-0.63

POSITIVE LOGITS

 emot

0.88

creen

0.84

 hello

0.84

 smile

0.80

 emoji

0.77

ggle

0.76

heet

0.75

 goodbye

0.73

olson

0.72

ys

0.72

Activations Density 0.021%

references to smiles

references to smiles and expressions of happiness

No Comments

No Known Activations