INDEX

Explanations

calories-related terms

oai_token-act-pair · gpt-3.5-turbo

references to calorie counts and dietary information

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

DragonMagazine

-0.81

smanship

-0.72

krit

-0.70

ker

-0.68

eer

-0.67

idable

-0.65

 Stellar

-0.65

unci

-0.64

sight

-0.64

acular

-0.63

POSITIVE LOGITS

 calories

1.25

ritional

1.17

rition

1.04

 calorie

1.01

 Calories

0.92

 intake

0.88

orie

0.87

 caloric

0.87

aceous

0.86

 metabolism

0.85

Activations Density 0.013%

calories-related terms

references to calorie counts and dietary information

No Comments

No Known Activations