INDEX

Explanations

references to error handling and fatal errors in code

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

test method calls on `t`

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

 queſta

-0.85

+#+#

-0.84

tvguidetime

-0.81

 betweenstory

-0.78

 kaarangay

-0.78

webElementXpaths

-0.78

Tikang

-0.78

niſſe

-0.77

ThroughAttribute

-0.77

الإنجليزية

-0.77

POSITIVE LOGITS

 XCTest

0.33

XCT

0.32

TestCase

0.31

JUnit

0.30

 dividido

0.28

 pomy

0.26

 wrong

0.26

 porté

0.25

 Pandemie

0.25

 TestCase

0.25

Activations Density 0.005%

references to error handling and fatal errors in code

test method calls on `t`

No Comments

No Known Activations

references to error handling and fatal errors in code

test method calls on `t`

No Comments

No Known Activations