INDEX

Explanations

complex sequences involving verbs and ownership

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

ThroughAttribute

-0.55

ControllerTest

-0.48

umna

-0.48

urier

-0.47

tieg

-0.47

omys

-0.46

Indexes

-0.46

もし

-0.46

BorderStyle

-0.45

 garam

-0.43

POSITIVE LOGITS

 himself

1.01

 herself

0.87

 yourselves

0.86

 yourself

0.83

 themselves

0.82

himself

0.81

ScopeManager

0.80

 ourselves

0.80

PrototypeOf

0.78

 Himself

0.75

Activations Density 0.715%

complex sequences involving verbs and ownership

No Comments

No Known Activations