INDEX

Explanations

keywords or significant terms related to various topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 дописавши

-0.57

-0.50

🇶

-0.47

 capital

-0.46

ustain

-0.46

 Capital

-0.46

-0.45

 aproxim

-0.43

 apparent

-0.43

 ostavi

-0.43

POSITIVE LOGITS

InjectAttribute

0.88

ThroughAttribute

0.77

writeField

0.74

 Theſe

0.73

 للاسماء

0.72

)».

0.72

 nakalista

0.71

 Houſe

0.71

 useAppContext

0.71

 الحره

0.68

Activations Density 2.578%

keywords or significant terms related to various topics

No Comments

No Known Activations