INDEX

Explanations

terms related to changes in text or documents

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

omitted or deleted text

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

ConfigureAwait

-0.42

ngdoc

-0.40

doGet

-0.39

 rainha

-0.35

-0.34

SuppressWarnings

-0.31

 blessed

-0.31

+#+

-0.30

 prêt

-0.30

 hiện

-0.30

POSITIVE LOGITS

deletion

2.88

 deletion

1.51

 Deletion

1.42

Deletion

1.40

 deletions

1.13

deleting

0.90

 deleting

0.89

PerformLayout

0.88

 deletes

0.88

 delete

0.83

Activations Density 0.000%

terms related to changes in text or documents

omitted or deleted text

No Comments

No Known Activations