INDEX

Explanations

phrases that express denial or refutation

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

DockStyle

-0.79

+#+#

-0.67

tagHelperRunner

-0.60

 Biôgrafia

-0.60

ConstraintMaker

-0.59

الإنجليزية

-0.58

tagext

-0.57

 transférez

-0.56

lgari

-0.56

BeginContext

-0.54

POSITIVE LOGITS

def

0.40

 البر

0.38

paddingVertical

0.38

 denied

0.38

def

0.38

 Ingram

0.37

denied

0.36

getInteger

0.36

 invalid

0.36

yargs

0.35

Activations Density 0.125%

phrases that express denial or refutation

No Comments

No Known Activations