INDEX

Explanations

mentions of the word "fat."

oai_token-act-pair · gpt-3.5-turbo

references to dietary fat

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Extend

-0.77

Atl

-0.76

 Puzzles

-0.75

boxes

-0.71

uden

-0.70

 Weaver

-0.69

akura

-0.67

outheast

-0.67

craft

-0.66

Atl

-0.65

POSITIVE LOGITS

fat

3.72

fat

2.71

Fat

2.44

Fat

2.35

 fats

2.19

FAT

1.99

 fatty

1.88

 adip

1.63

 obese

1.47

 obesity

1.31

Activations Density 0.028%

mentions of the word "fat."

references to dietary fat

No Comments

No Known Activations