INDEX

Explanations

references to specific technology companies and their products

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

tech company services

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

fromnode

-0.56

 BrowserModule

-0.52

Tikang

-0.52

 näytte

-0.52

RegressionTest

-0.50

EndContext

-0.47

曖昧さ回避

-0.47

̈́

-0.46

 ujednoznacz

-0.45

 malheureux

-0.45

POSITIVE LOGITS

 Cloud

0.51

ANDUM

0.51

 dime

0.49

DNA

0.47

 cloud

0.47

 digital

0.47

VolleyError

0.46

 Citizen

0.45

 Lilly

0.45

 Volt

0.45

Activations Density 0.035%

references to specific technology companies and their products

tech company services

No Comments

No Known Activations