INDEX

Explanations

phrases related to high levels or concentrations in various contexts

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

adjectival intensity descriptors indicating a high level, especially when used as hyphenated compound modifiers, and occasionally their contrasting opposite-extreme forms.

oai_token-act-pair · gpt-5 Triggered by @liuliujiujiu05

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_11/width_16k/average_l0_80

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

adaptiveStyles

-0.74

 Trotter

-0.72

зец

-0.70

 voyez

-0.70

UserAgent

-0.70

 redé

-0.68

 entière

-0.68

 Schäfer

-0.68

 Zelanda

-0.68

 Creatures

-0.67

POSITIVE LOGITS

 High

1.56

 high

1.51

High

1.49

 HIGH

1.47

high

1.40

HIGH

1.39

Low

1.22

Low

1.16

高

1.12

low

1.11

Activations Density 0.141%

phrases related to high levels or concentrations in various contexts

adjectival intensity descriptors indicating a high level, especially when used as hyphenated compound modifiers, and occasionally their contrasting opposite-extreme forms.

No Comments

No Known Activations