INDEX

Explanations

food-related terms, especially those related to health and nutrition

oai_token-act-pair · gpt-3.5-turbo

references to food items and their qualities or effects

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

 Dhabi

-0.80

Ã±o

-0.78

 Riders

-0.73

 Blades

-0.72

 Bots

-0.68

 Drone

-0.67

-0.66

 Stra

-0.66

vernment

-0.65

vu

-0.65

POSITIVE LOGITS

 eaten

1.07

 foods

1.05

 consumed

0.90

Ingredients

0.89

 additives

0.88

 allergies

0.87

 ingested

0.87

 eater

0.86

 nutrit

0.85

paste

0.85

Activations Density 0.031%

food-related terms, especially those related to health and nutrition

references to food items and their qualities or effects

No Comments

No Known Activations