INDEX

Explanations

phrases indicating significant impacts or consequences of events or actions

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

 bort

-0.51

 Alternative

-0.46

Alternative

-0.46

La

-0.45

сен

-0.45

liez

-0.43

 McAllister

-0.43

ams

-0.43

 alternative

-0.42

 perfec

-0.42

POSITIVE LOGITS

 impact

1.66

impact

1.54

 impacts

1.50

Impact

1.45

 effects

1.45

 Impact

1.42

 effect

1.40

Impacts

1.37

effect

1.36

 Impacts

1.36

Activations Density 0.489%

phrases indicating significant impacts or consequences of events or actions

No Comments

No Known Activations