INDEX

Explanations

elements and functions related to testing in programming

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

org.apache.commons.math

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

 Chwiliwch

-1.05

WireFormatLite

-0.85

+#+#

-0.79

ſelf

-0.76

 ſta

-0.75

CloseOperation

-0.72

 queſta

-0.72

MLLoader

-0.72

 Houſe

-0.71

Efq

-0.71

POSITIVE LOGITS

 simple

0.38

 получилось

0.35

du

0.34

 Travis

0.34

 point

0.34

su

0.33

 basic

0.33

 using

0.32

 slou

0.32

 Bischof

0.31

Activations Density 0.021%

elements and functions related to testing in programming

org.apache.commons.math

No Comments

No Known Activations