INDEX

Explanations

phrases that describe different categories or classifications of types or kinds

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

this type of

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

IBOutlet

-0.58

SEGUIR

-0.47

 ContentValues

-0.47

 propOrder

-0.46

Owners

-0.45

 Fairness

-0.45

ioutil

-0.45

LookAnd

-0.45

 réfrig

-0.45

 raiſ

-0.44

POSITIVE LOGITS



0.58

此

0.52

 denne

0.51

こういう

0.51

 dieser

0.48

 acestei

0.46

这类

0.45

 daqui

0.43

 Cette

0.43

 denna

0.43

Activations Density 0.017%

phrases that describe different categories or classifications of types or kinds

this type of

No Comments

No Known Activations