INDEX

Explanations

specific metadata and identifiers related to records or files

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

records, serial numbers, references

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

awtextra

-0.63

 onOptions

-0.58

RegressionTest

-0.51

 nahilalakip

-0.50

 initComponents

-0.50

 cherchés

-0.49

__(/*!

-0.49

StoryboardSegue

-0.46

addContainerGap

-0.45

RegistryLite

-0.45

POSITIVE LOGITS

ollectionView

0.47

 Bergh

0.44

мато

0.40

+#+#

0.40

Lum

0.40

Fou

0.39

Cop

0.39

tea

0.38

Fou

0.38

Seq

0.38

Activations Density 0.084%

specific metadata and identifiers related to records or files

records, serial numbers, references

No Comments

No Known Activations