INDEX

Explanations

the name "David" with varying strengths of activation

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "David" in various contexts

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

cffffcc

-1.03

NetMessage

-0.86

ĺħ

-0.80

Reply

-0.77

phies

-0.76

ttes

-0.74

isode

-0.73

¥ŀ

-0.72

inventoryQuantity

-0.72

urgical

-0.71

POSITIVE LOGITS

 Bowie

1.10

 Cameron

0.97

 Axel

0.96

 Beckham

0.94

 Petraeus

0.90

Ic

0.90

 Rockefeller

0.89

son

0.86

 Lynch

0.83

 Kore

0.82

Activations Density 0.027%

the name "David" with varying strengths of activation

mentions of the name "David" in various contexts

No Comments

No Known Activations