INDEX

Explanations

The neuron is looking for words related to expressions of affection

oai_token-act-pair · gpt-3.5-turbo

expressions of affection and fondness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ulhu

-0.92

ÄŁ

-0.76

ramid

-0.74

akedown

-0.71

soDeliveryDate

-0.70

krit

-0.65

medi

-0.65

ozo

-0.63

proof

-0.62

DoS

-0.62

POSITIVE LOGITS

 affection

1.00

ately

0.98

 fond

0.83

 kisses

0.76

ate

0.73

uously

0.72

0.70

atile

0.68

 affinity

0.68

 passionately

0.67

Activations Density 0.059%

The neuron is looking for words related to expressions of affection

expressions of affection and fondness

No Comments

No Known Activations