INDEX

Explanations

key terms related to data analysis and tracking in scientific contexts

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

specialized terms

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

in

-0.38

and

-0.37

by

-0.37

 latest

-0.36

to

-0.36

 facing

-0.34

 geométrico

-0.34

 with

-0.34

 pār

-0.33

di

-0.32

POSITIVE LOGITS

UserScript

0.81

IsMutable

0.81

 препратки

0.76

rungsseite

0.72



0.70

 パンチラ

0.68

<unused80>

0.65

<unused43>

0.65

<unused8>

0.65

[@BOS@]

0.65

Activations Density 0.177%

key terms related to data analysis and tracking in scientific contexts

specialized terms

No Comments

No Known Activations