INDEX

Explanations

elements indicative of technical or programming contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

Illuminate

-0.60

JpaRepository

-0.55

webElementXpaths

-0.49

-0.47

 Illuminate

-0.47

)(((

-0.44

Diweddarwch

-0.43

jface

-0.43

 aDecoder

-0.41

 anda

-0.41

POSITIVE LOGITS

UnsafeEnabled

1.23

الإنجليزية

1.16

 faſt

1.09

Efq

1.09

 myſelf

1.06

 purpoſe

1.03

 ſta

1.02

ſelves

1.01

 ſtre

1.01

✨:

1.00

Activations Density 2.504%

elements indicative of technical or programming contexts

No Comments

No Known Activations