INDEX

Explanations

sequences of whitespace or formatting markers in code

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

programming code snippets

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

AnimationsModule

-0.60

WriteBarrier

-0.54

CppMethod

-0.54

TintMode

-0.53

Демографія

-0.48

 esternos

-0.47

IntoConstraints

-0.47

 arşivlendi

-0.47

classnames

-0.47

 оригіналу

-0.46

POSITIVE LOGITS

AlterField

0.42

Personendaten

0.38

 GenerationType

0.37

WriteTagHelper

0.37

ser

0.36

########.

0.36

 paci

0.35

 contr

0.35

Params

0.34

bio

0.34

Activations Density 0.005%

sequences of whitespace or formatting markers in code

programming code snippets

No Comments

No Known Activations