INDEX

Explanations

references to programming or code-related components

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

Code, URLs, and file paths

np_acts-logits-general · gemini-2.0-flash

file, views, court, pharma

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

in

-0.65

–

-0.65

by

-0.63

has

-0.62

is

-0.61

-0.60

at

-0.59

POSITIVE LOGITS

Portail

0.88

 Theſe

0.86

common

0.85

 otomatig

0.84

 Reſ

0.83

expandindo

0.79

 utafitiHapana

0.78

 {}));

0.78

enumi

0.78

Activations Density 0.114%

references to programming or code-related components

Code, URLs, and file paths

file, views, court, pharma

No Comments

No Known Activations