INDEX

Explanations

specific product names or identifiers related to technology or computer hardware

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

end of turn

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AutoScaleMode

-0.47

Geplaatst

-0.44

XmlAccessType

-0.42

))).

-0.41

DebuggerStep

-0.41

enumii

-0.41

RTLR

-0.40

Jîn

-0.39

sizeCache

-0.39

Autoritní

-0.39

POSITIVE LOGITS

uginosa

0.55

gridColumn

0.54

bootstrapcdn

0.53

acheco

0.52

 يتيمه

0.52

shmi

0.51

 dieſe

0.50

loroethane

0.50

ⓧ

0.49

OOTDTY

0.48

Activations Density 0.050%

specific product names or identifiers related to technology or computer hardware

end of turn

No Comments

No Known Activations