INDEX

Explanations

patterns of asterisks or wildcard symbols

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

documentation and parameter descriptions

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

parsedMessage

-1.11

 Paglinawan

-0.91

IntoConstraints

-0.91

featureID

-0.89

PerformLayout

-0.88

 MessageBoxIcon

-0.88

awtextra

-0.86

 nakalista

-0.85

 pinulongan

-0.84

principalColumn

-0.84

POSITIVE LOGITS

↵

0.38

IC

0.38

IP

0.32

PLIC

0.32

 known

0.32

 Mark

0.32

ED

0.32

ING

0.32

Mark

0.32

 This

0.31

Activations Density 0.005%

patterns of asterisks or wildcard symbols

documentation and parameter descriptions

No Comments

No Known Activations