INDEX

Explanations

terms related to cleaning agents and methods

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

Newly

-0.40

 Newly

-0.38

 solidly

-0.36

 lành

-0.35

KeyValuePair

-0.35

 reckoned

-0.35

 saja

-0.34

 Oskar

-0.34

aneamente

-0.34

 barn

-0.34

POSITIVE LOGITS

MLLoader

0.64

CloseOperation

0.59

:✨

0.56

 zijne

0.55

aarrggbb

0.53

 Administrativna

0.52

ьаж

0.50

]-->

0.48

 pinulongan

0.48

RectangleBorder

0.48

Activations Density 0.007%

terms related to cleaning agents and methods

No Comments

No Known Activations