INDEX

Explanations

concepts related to self-defense and protective actions

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

bootstrapcdn

-0.58

 lenker

-0.49

mapStateToProps

-0.44

下一个

-0.39

 megfe

-0.38

-0.36

BindingSource

-0.36

getDescriptor

-0.36

XtraReports

-0.35

RunWith

-0.35

POSITIVE LOGITS

 protecting

0.64

 protect

0.61

 protection

0.60

 indignation

0.59

protect

0.59

PROTECT

0.58

 defense

0.57

捍

0.55

 defend

0.55

 défendre

0.54

Activations Density 0.420%

concepts related to self-defense and protective actions

No Comments

No Known Activations