INDEX

Explanations

characters or symbols commonly used in coding or syntax

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

programming test frameworks

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

httphttps

-0.60

TagMode

-0.52

 FetchType

-0.51

eqn

-0.50

findpost

-0.47

новништво

-0.46

 Meksiku

-0.46

:✨

-0.45

="@+

-0.45

EndContext

-0.41

POSITIVE LOGITS

IUrlHelper

0.60

RSpec

0.53

ViewFeatures

0.50

scalatest

0.48

RunWith

0.45

 OnInit

0.44

Autowired

0.44

Materiał

0.43

testify

0.43

magem

0.43

Activations Density 0.000%

characters or symbols commonly used in coding or syntax

programming test frameworks

No Comments

No Known Activations