INDEX

Explanations

discussion of measurement outcomes and variables

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

cydow

-0.35

 mieście

-0.33

 Züge

-0.32

asanjo

-0.31

gql

-0.31

 Regeln

-0.31

Begriffsklä

-0.31

amssymb

-0.29

 tény

-0.29

 zdecydow

-0.29

POSITIVE LOGITS

PerformLayout

0.77

LookAnd

0.72

styleType

0.68

 Italijani

0.62

titleMargin

0.62

 Italijanski

0.59

expandindo

0.56

↹

0.55

aarrggbb

0.54

uxxxx

0.53

Activations Density 0.819%

discussion of measurement outcomes and variables

No Comments

No Known Activations