INDEX

Explanations

patterns and structures in mathematical or programming contexts

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

code and mathematical notation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

GEBURTSDATUM

-0.60

 IFTT

-0.58

 typelib

-0.49

jspb

-0.47

LookAnd

-0.46

RegressionTest

-0.44

 ✭✭

-0.44

windowFixed

-0.44

Бахар

-0.44

 Italijanski

-0.43

POSITIVE LOGITS

 everywhere

0.39

principalTable

0.38

ospels

0.36

 genü

0.36

 elsewhere

0.36

HtmlAttribute

0.35

 Everywhere

0.35

MENAFN

0.35

 again

0.35

 насељу

0.34

Activations Density 0.029%

patterns and structures in mathematical or programming contexts

code and mathematical notation

No Comments

No Known Activations