INDEX

Explanations

concepts related to academic discourse and theoretical frameworks

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

new interpretation or theory

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

+#+

-0.45

 defaultstate

-0.45

 FetchType

-0.43

 GenerationType

-0.43

Diwedd

-0.41

าสตร์

-0.40

 fian

-0.40

RTEE

-0.38

KommentareTeilen

-0.38

 Aqueous

-0.37

POSITIVE LOGITS

 وتسجيلات

0.52

 methodological

0.46

SuppressLint

0.44

 hitherto

0.44

 sociologists

0.44

 repens

0.43

 policymakers

0.43

uzn

0.42

 privilegi

0.42

 scholars

0.42

Activations Density 0.052%

concepts related to academic discourse and theoretical frameworks

new interpretation or theory

No Comments

No Known Activations