INDEX

Explanations

sequences of characters that match specific coding syntax or identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

formik

-0.60

 preocupes

-0.56

 pungkasnya

-0.53

 Multipart

-0.52

зулта

-0.51

AsyncTask

-0.51

Best

-0.50

 mahdollis

-0.50

 ogóle

-0.49

varargin

-0.49

POSITIVE LOGITS

 BoxDecoration

0.58

0.57

IonicModule

0.54

paravant

0.53

anthin

0.52

&=&\

0.50

leaved

0.48

rophes

0.48

ricula

0.48

tify

0.47

Activations Density 1.354%

sequences of characters that match specific coding syntax or identifiers

No Comments

No Known Activations