INDEX

Explanations

elements related to programming or assembly attributes

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 propOrder

-0.61

 houſe

-0.59

SerializedName

-0.57

 ſche

-0.57

 ſever

-0.56

 leſs

-0.56

ſelf

-0.56

 myſelf

-0.55

 ſch

-0.55

 Autorisations

-0.55

POSITIVE LOGITS

0.60

an

0.52

 Wikimedijinoj

0.50

fjspx

0.45

BorderFactory

0.43

__(/*!

0.41

 regular

0.41

\}=\

0.38

pa

0.38

)}=\

0.37

Activations Density 0.357%

elements related to programming or assembly attributes

No Comments

No Known Activations