INDEX

Explanations

symbols and characters related to code and programming structures

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

user

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

writeFieldEnd

-0.58

SharedCtor

-0.53

:✨

-0.47

 onCreateView

-0.42

 <>",

-0.40

Xna

-0.40

енча

-0.40

awaiter

-0.40

reathe

-0.38

 OFDb

-0.38

POSITIVE LOGITS

 Vikipedi

0.47

앵

0.46

IUrlHelper

0.44

PMailer

0.43

новништво

0.43

 poshmark

0.43

XtraReports

0.43

 dieß

0.42

Интер

0.40

 múl

0.39

Activations Density 0.002%

symbols and characters related to code and programming structures

user

No Comments

No Known Activations