INDEX

Explanations

temporal references and descriptions of experiences

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

sequence follow-up

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

 ModelExpression

-0.57

IntoConstraints

-0.53

 Италијани

-0.52

 ProtoMessage

-0.51

ConstraintMaker

-0.51

Personendaten

-0.50

 forState

-0.48

GEBURTS

-0.48

 localObject

-0.47

httphttps

-0.47

POSITIVE LOGITS

 rematch

0.41

ẵng

0.36

acchar

0.35

 ActionBar

0.35

السكان

0.34

 adjournment

0.33

sequent

0.33

umblr

0.33

 Subsequent

0.33

Спасылкі

0.32

Activations Density 0.134%

temporal references and descriptions of experiences

sequence follow-up

No Comments

No Known Activations