INDEX

Explanations

patterns in code related to string manipulation and replacements

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

replace or replace with

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 Obrador

-0.44

Personensuche

-0.41

LabelTagHelper

-0.41

MarshalTo

-0.40

 ProtoMessage

-0.39

 estimés

-0.39

 groupName

-0.36

dames

-0.35

点此举报

-0.35

pozy

-0.35

POSITIVE LOGITS

 replace

0.62

 Paglinawan

0.54

 replacement

0.54

 replacing

0.53

replace

0.52

Replace

0.52

 Replace

0.50

MessageOf

0.50

 replacements

0.49

 replaced

0.48

Activations Density 0.010%

patterns in code related to string manipulation and replacements

replace or replace with

No Comments

No Known Activations