INDEX

Explanations

data and findings related to research outcomes and analyses

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

reports, findings, or complaints

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

 FontFamily

-0.44

-0.42

 Commercial

-0.41

 Transportation

-0.41

infinite

-0.40

 main

-0.39

 Main

-0.38

 imaginary

-0.38

 Ishikawa

-0.38

POSITIVE LOGITS

 nahilalakip

0.71

AnchorStyles

0.62

SequentialGroup

0.58

hyrchwyd

0.57

MessageTagHelper

0.55

 PeEnEo

0.54

rawDesc

0.53

 оригіналу

0.52

 consultato

0.52

VIAF

0.51

Activations Density 0.087%

data and findings related to research outcomes and analyses

reports, findings, or complaints

No Comments

No Known Activations