INDEX

Explanations

specific technical terms related to scanning and data processing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

}".

-0.43

dest

-0.39

\}\\

-0.38

}'.

-0.36

ysty

-0.36

}]

-0.35

ges

-0.34

']],

-0.34

...

-0.34

POSITIVE LOGITS

an

0.79

anners

0.75

Gan

0.75

 Doran

0.74

 Lancet

0.72

nan

0.72

Gan

0.71

ana

0.71

lan

0.71

Lan

0.71

Activations Density 0.737%

specific technical terms related to scanning and data processing

No Comments

No Known Activations