INDEX

Explanations

elements of structural or mathematical notation

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

mathematical hat notation

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

PerformLayout

-0.74

 wireType

-0.71

RegressionTest

-0.67

ChildScrollView

-0.63

IntoConstraints

-0.63

 propOrder

-0.62

 HasFactory

-0.60

 ſal

-0.59

outSlope

-0.59

 resourceCulture

-0.58

POSITIVE LOGITS

om

0.38

*}[

0.35

downcase

0.35

hausen

0.33

0.32

료

0.32

extern

0.32

EN

0.31

分

0.31

 getActivity

0.31

Activations Density 0.135%

elements of structural or mathematical notation

mathematical hat notation

No Comments

No Known Activations