INDEX

Explanations

details related to scientific observations and evidence

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

ViewFeatures

-0.70

tvguidetime

-0.65

!*\

-0.64

AndEndTag

-0.64

IntoConstraints

-0.61

 }}"></

-0.61

 betweenstory

-0.60

CloseOperation

-0.58

 estekak

-0.58

verifyException

-0.54

POSITIVE LOGITS

 raises

1.75

 raise

1.52

 suggests

1.43

 raising

1.42

 Raises

1.40

 Raise

1.35

 suggesting

1.35

 suggest

1.33

 raised

1.32

raises

1.29

Activations Density 0.700%

details related to scientific observations and evidence

No Comments

No Known Activations