INDEX

Explanations

technical terminology and data structures related to programming and circuits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 otomatig

-0.40

ceğine

-0.39

ⓧ

-0.38

 snippetHide

-0.37

 дописавши

-0.35

 lagi

-0.35

WriteBarrier

-0.33

 bang

-0.33

 '\\;'

-0.31

IndentedString

-0.31

POSITIVE LOGITS

iverr

0.45

ကိုးကား

0.44

URBANA

0.44

ſelves

0.43

 juſt

0.43

ธ์

0.43

 diſt

0.42

 voet

0.42

 viſ

0.42

 ſou

0.41

Activations Density 0.753%

technical terminology and data structures related to programming and circuits

No Comments

No Known Activations