INDEX

Explanations

consistent use of the word "always"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 17-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_17/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.17.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

出版年

-0.74

RuleContext

-0.68

reportWebVitals

-0.67

Witam

-0.66

Datuak

-0.64

ioutil

-0.64

 Bjør

-0.62

er

-0.61

mazoo

-0.61

 IDisposable

-0.60

POSITIVE LOGITS

 ALWAYS

0.87

 Always

0.81

 always

0.78

Always

0.78

ALWAYS

0.76

always

0.76

 luôn

0.74

しも

0.72

awsze

0.67

sidemargin

0.65

Activations Density 0.081%

consistent use of the word "always"

No Comments

No Known Activations