INDEX

Explanations

phrases centered around focus and attention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 8-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_8/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.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

 Methuen

-0.85

Plata

-0.70

Pip

-0.67

lblNewLabel

-0.66

 Schalke

-0.65

 tubing

-0.65

 Ingram

-0.65

nb

-0.64

umato

-0.63

;/

-0.62

POSITIVE LOGITS

 focus

1.96

FOCUS

1.92

 FOCUS

1.84

 Focus

1.81

Focus

1.73

focus

1.72

 Fokus

1.67

 focuses

1.64

 focussed

1.62

 focused

1.57

Activations Density 0.081%

phrases centered around focus and attention

No Comments

No Known Activations