INDEX

Explanations

references to individuals in leadership roles or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Gew

-0.73

SafeAreaView

-0.73

 Hül

-0.67

 Kenn

-0.65

ikkert

-0.63

liven

-0.63

Lep

-0.63

 Cunha

-0.62

ouds

-0.62

ꜱ

-0.62

POSITIVE LOGITS

 directors

2.12

 director

2.08

director

1.96

 Director

1.93

 Directors

1.92

 DIRECTOR

1.92

Director

1.80

Directors

1.80

directors

1.73

DIRECTOR

1.70

Activations Density 0.049%

references to individuals in leadership roles or titles

No Comments

No Known Activations