INDEX

Explanations

negative expressions and forms of denial

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

intios

-0.53

 Chwiliwch

-0.39

 vertrouwen

-0.38

 pinulongan

-0.38

 gezegd

-0.36

 anglès

-0.36

 '-';

-0.36

 ffilmiau

-0.35

 Ahnung

-0.35

 unſ

-0.35

POSITIVE LOGITS

ьаж

0.50

MockBean

0.49

MLLoader

0.47

UserScript

0.44

httphttps

0.43

endphp

0.43

BASEPATH

0.42

endpush

0.42

isNameExpr

0.41

FromNib

0.40

Activations Density 0.278%

negative expressions and forms of denial

No Comments

No Known Activations