INDEX

Explanations

mentions of the company "Warner Bros" with varying emphasis

oai_token-act-pair · gpt-3.5-turbo

references to Warner Bros and associated entities

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

aram

-0.72

etus

-0.70

orem

-0.65

acly

-0.64

natureconservancy

-0.63

pod

-0.61

 liner

-0.61

phis

-0.61

yll

-0.59

ypes

-0.59

POSITIVE LOGITS

 Bros

1.03

 Entertainment

0.89

 Stores

0.88

 Pictures

0.88

 Interactive

0.84

 Animation

0.81

 Studios

0.79

Ltd

0.76

 Television

0.75

 Brothers

0.75

Activations Density 0.033%

mentions of the company "Warner Bros" with varying emphasis

references to Warner Bros and associated entities

No Comments

No Known Activations