INDEX

Explanations

technical terms and measurements related to radio frequencies and signal processing

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

index, value, measurement, ranges

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

AddTagHelper

-0.83

RenderAtEndOf

-0.77

IntoConstraints

-0.76

tonode

-0.71

IsMutable

-0.70

ArrowToggle

-0.69

expandindo

-0.67

 gynhyrchwyd

-0.67

imageshack

-0.67

ReusableCell

-0.67

POSITIVE LOGITS

 ranging

0.36

 ranges

0.31

 range

0.31

 alcanzado

0.30

 levels

0.30

 alcanzar

0.29

low

0.29

 رک

0.29

 nég

0.29

 value

0.27

Activations Density 0.250%

technical terms and measurements related to radio frequencies and signal processing

index, value, measurement, ranges

No Comments

No Known Activations