INDEX

Explanations

references to the emotion "happy."

oai_token-act-pair · gpt-3.5-turbo

expressions of happiness and related emotions

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

ngth

-0.83

soDeliveryDate

-0.77

arin

-0.71

atum

-0.71

çīĪ

-0.70

oug

-0.70

essential

-0.69

ioxide

-0.68

ciplinary

-0.68

DoS

-0.68

POSITIVE LOGITS

joy

0.83

vale

0.76

istic

0.72

 endings

0.72

 happily

0.70

 Meal

0.66

sticks

0.66

 âĶľ

0.64

 Pupp

0.62

 happy

0.62

Activations Density 0.037%

references to the emotion "happy."

expressions of happiness and related emotions

No Comments

No Known Activations