INDEX

Explanations

specific references to mathematical objects and symbols

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

equations, figures, or tables

np_acts-logits-general · gemini-2.0-flash

scientific names and symbols

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 Tind

-0.89

 Hart

-0.72

Hart

-0.71

Mee

-0.71

 Hutt

-0.69

 Knox

-0.68

 Salamanca

-0.68

 Ferri

-0.68

Nak

-0.67

 Perrin

-0.66

POSITIVE LOGITS

 Nike

0.75

 DuPont

0.74

 Kane

0.72

HP

0.70

Nike

0.69

 Harry

0.68

#+#

0.68

Kane

0.68

Ka

0.66

 Levit

0.65

Activations Density 0.798%

specific references to mathematical objects and symbols

equations, figures, or tables

scientific names and symbols

No Comments

No Known Activations