INDEX

Explanations

terms and phrases related to encryption and cryptographic algorithms

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

enumi

-0.58

 unknownFields

-0.58

hdashline

-0.52

createServer

-0.51

publicPath

-0.51

]--;

-0.50

-0.49

 EClass

-0.49

 surla

-0.47

itize

-0.47

POSITIVE LOGITS

padd

0.65

 window

0.63

 block

0.61

window

0.60

 shift

0.58

 Block

0.57

 оригіналу

0.55

 WINDOW

0.54

 Window

0.54

 BLOCK

0.54

Activations Density 0.439%

terms and phrases related to encryption and cryptographic algorithms

No Comments

No Known Activations