INDEX

Explanations

JavaScript operations related to deleting or unsetting properties and objects

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

delete or unset

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

 queſta

-0.74

ollectionView

-0.69

<unused41>

-0.60

<unused8>

-0.60

<unused47>

-0.59

<unused28>

-0.59

<unused74>

-0.59

<unused79>

-0.59

[@BOS@]

-0.59

<unused14>

-0.59

POSITIVE LOGITS

unset

0.69

IVEREF

0.56

RemoveField

0.51

 removed

0.41

removeAttr

0.40

abestanden

0.40

removeAttribute

0.39

setopt

0.39

hdysval

0.39

 unset

0.38

Activations Density 0.003%

JavaScript operations related to deleting or unsetting properties and objects

delete or unset

No Comments

No Known Activations