INDEX

Explanations

references to physical body weight, weight loss, weight gain, body image, and associated medical terms

oai_token-act-pair · gpt-3.5-turbo

references to body weight and weight-related concerns

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

 dispatch

-0.77

Roe

-0.72

sonian

-0.70

Pon

-0.69

 dispatched

-0.69

NRS

-0.68

 Morse

-0.68

 RAND

-0.68

 Administ

-0.67

oral

-0.67

POSITIVE LOGITS

 physique

1.01

 hormones

0.92

fat

0.89

 dysph

0.87

 hormone

0.81

 cortisol

0.81

 athlet

0.80

 calves

0.80

agen

0.79

 disorders

0.78

Activations Density 0.085%

references to physical body weight, weight loss, weight gain, body image, and associated medical terms

references to body weight and weight-related concerns

No Comments

No Known Activations