INDEX

Explanations

references to mathematical operations and tests within code

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

test method names

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

AndEndTag

-0.60

aarrggbb

-0.60

 strerror

-0.56

 queſta

-0.56

tagHelperRunner

-0.53

 Meksiku

-0.52

RegressionTest

-0.52

ſelves

-0.51

URLException

-0.51

 cherchés

-0.51

POSITIVE LOGITS

ScopeManager

0.58

bas

0.35

 GenerationType

0.35

maxcdn

0.33

Viitteet

0.33

 вовсе

0.33

ÁF

0.33

Har

0.33

af

0.32

Mc

0.32

Activations Density 0.053%

references to mathematical operations and tests within code

test method names

No Comments

No Known Activations