INDEX

Explanations

mentions of the word "defense" and its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 发表于

-0.58

isShowing

-0.57

RegressionTest

-0.55

 نخ

-0.51

lepiej

-0.50

 bislang

-0.50

 bono

-0.49

 Justo

-0.49

gating

-0.49

 Kalt

-0.48

POSITIVE LOGITS

AndEndTag

0.88

 lenker

0.71

GEBURTSDATUM

0.69

 autorytatywna

0.56

 nahilalakip

0.55

 المعيارى

0.54

AutoresizingMask

0.53

ragan

0.52

 transfieras

0.51

argout

0.51

Activations Density 0.076%

mentions of the word "defense" and its variations

No Comments

No Known Activations