INDEX

Explanations

references to statistical analysis and experimental studies

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

WithIOException

-0.45

 taas

-0.41

Segue

-0.36

トイ

-0.35

 Hays

-0.34

 multitud

-0.33

 ángeles

-0.33

 various

-0.33

жество

-0.33

 bread

-0.33

POSITIVE LOGITS

 nakalista

0.75

ScopeManager

0.65

 يتيمه

0.57

ItemBackground

0.55

исленность

0.50

SharedDtor

0.49

Obrázky

0.48

 gynhyrchwyd

0.48

 Audiodateien

0.47

Hentet

0.46

Activations Density 0.683%

references to statistical analysis and experimental studies

No Comments

No Known Activations