INDEX

Explanations

expressions of doubt or skepticism about statements and ideas

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

exactly, surprisingly

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

Atsauces

-0.48

ScopeManager

-0.48

 encuentras

-0.46

 casó

-0.43

MLLoader

-0.43

 régulier

-0.43

 réguli

-0.42

 cuerdas

-0.41

 abstrait

-0.40

 CreateTagHelper

-0.40

POSITIVE LOGITS

rungsseite

0.56

neb

0.50

ombus

0.50

 yeah

0.48

 jTextField

0.47

Yeah

0.46

dame

0.46

yeah

0.45

 Yeah

0.45

cycline

0.45

Activations Density 0.061%

expressions of doubt or skepticism about statements and ideas

exactly, surprisingly

No Comments

No Known Activations