INDEX

Explanations

mentions of calorie-related information

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 @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Breach

-0.73

hawk

-0.73

 Pathfinder

-0.70

 Quinn

-0.69

Phoenix

-0.67

king

-0.67

 Construct

-0.65

Uk

-0.65

 Phantom

-0.65

opter

-0.65

POSITIVE LOGITS

 calories

3.79

 calorie

2.69

 Calories

2.58

 caloric

2.28

 kcal

1.92

 carbs

1.76

 carbohydrates

1.61

 nutrition

1.49

 vitamins

1.43

 nutritional

1.42

Activations Density 0.010%

mentions of calorie-related information

references to calorie counts and dietary information

No Comments

No Known Activations