INDEX

Explanations

the name "Trevor" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

references to the name "Trevor."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 discipl

-0.73

 actionGroup

-0.72

maid

-0.69

ĻĤ

-0.69

spring

-0.67

 circ

-0.66

arching

-0.66

 corros

-0.66

lesiastical

-0.64

 responsive

-0.63

POSITIVE LOGITS

 Trevor

1.41

 Noah

0.84

 Gardner

0.84

istas

0.83

 Philips

0.83

 Cummings

0.83

 Damon

0.81

 Reeves

0.80

 Hoffman

0.80

 Draper

0.79

Activations Density 0.007%

the name "Trevor" with varying levels of activation

references to the name "Trevor."

No Comments

No Known Activations