INDEX

Explanations

markers indicating the beginning of important sections or headings in the text

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

Code snippets or fragments

np_acts-logits-general · gemini-2.0-flash

code and data strings

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

 محفوظة

-0.56

rikes

-0.47

ArgumentParser

-0.46

плом

-0.45

WebpackPlugin

-0.45

tigma

-0.44

UrlResolution

-0.42

 Svensson

-0.42

Sockets

-0.42

 calendriers

-0.41

POSITIVE LOGITS

 kaarangay

0.62

 jadx

0.61

saurus

0.60

>{@

0.57

 ISNI

0.57

rigkeit

0.56

aarrggbb

0.56

 nakalista

0.56

iastes

0.56

ldc

0.55

Activations Density 0.090%

markers indicating the beginning of important sections or headings in the text

Code snippets or fragments

code and data strings

No Comments

No Known Activations