INDEX

Explanations

references to governmental or ministerial roles and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 препратки

-0.53

webElementXpaths

-0.52

⟬

-0.46

parsedMessage

-0.45

StartTag

-0.44

 getItemId

-0.43

ThroughAttribute

-0.43

TagMode

-0.41

RegressionTest

-0.40

tac

-0.40

POSITIVE LOGITS

 Minister

0.86

 Ministry

0.82

 minister

0.73

Minister

0.71

Ministry

0.68

 ministry

0.64

 وزارت

0.63

 Ministro

0.61

 ministre

0.61

minister

0.61

Activations Density 0.170%

references to governmental or ministerial roles and titles

No Comments

No Known Activations