INDEX

Explanations

references to the concept of 'twins'

oai_token-act-pair · gpt-3.5-turbo

references to twins or twin-related concepts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

anwhile

-0.87

CoC

-0.84

andise

-0.76

UME

-0.71

uddin

-0.67

ulkan

-0.66

ãģĵ

-0.65

pmwiki

-0.65

aining

-0.64

ktop

-0.64

POSITIVE LOGITS

ned

1.22

ning

1.18

fold

0.92

 towers

0.88

 brother

0.85

 peaks

0.82

xes

0.81

 sister

0.80

 Peaks

0.78

 brothers

0.77

Activations Density 0.031%

references to the concept of 'twins'

references to twins or twin-related concepts

No Comments

No Known Activations