INDEX

Explanations

punctuation marks, particularly semicolons

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

technical contexts

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

MockBean

-0.47

IUrlHelper

-0.45

 seyn

-0.43

 кӀош

-0.42

AnimationsModule

-0.41

 Irán

-0.40

 dentaire

-0.40

 bluzka

-0.39

 intérêts

-0.39

andExpect

-0.38

POSITIVE LOGITS

 MainAxisSize

0.56

Datuak

0.49

AndEndTag

0.49

ugal

0.45

 صوتيه

0.45

Autoritní

0.45

GeneratedMessage

0.45

 محفوظة

0.44

 فريبيس

0.44

awaiter

0.43

Activations Density 0.000%

punctuation marks, particularly semicolons

technical contexts

No Comments

No Known Activations