INDEX

Explanations

references to specific data structures or indices within a dataset

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

Hexadecimal codes

np_acts-logits-general · gemini-2.0-flash

idx followed by hex values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/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.20.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

findpost

-1.50

rungsseite

-1.49

 виправивши

-1.40

fjspx

-1.39

 propOrder

-1.35

windowFixed

-1.27

دانشنامهٔ

-1.26

 tartalomajánló

-1.24

 الحره

-1.24

 للاسماء

-1.24

POSITIVE LOGITS

0.47

0.46

0.44

0.42

0.41

liko

0.40

Activations Density 0.607%

references to specific data structures or indices within a dataset

Hexadecimal codes

idx followed by hex values

No Comments

No Known Activations