INDEX

Explanations

affirmations of belief or conviction

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

ar

-0.77

lot

-0.77

-0.73

-0.72

-0.70

 lots

-0.69

SE

-0.67

us

-0.66

quad

-0.66

-0.65

POSITIVE LOGITS

 BELIEVE

1.81

 believe

1.77

believe

1.75

 believes

1.70

 believed

1.67

Believe

1.61

 Believe

1.60

belie

1.58

 Belief

1.54

 Beliefs

1.54

Activations Density 0.077%

affirmations of belief or conviction

No Comments

No Known Activations