INDEX

Explanations

terms and concepts related to technical specifications and classifications

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

semiparametric modeling and technical code snippets

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

 besök

-0.36

 équilibr

-0.35

STANDING

-0.35

 gambe

-0.34

いかがでしたか

-0.33

 raad

-0.32

 étr

-0.32

 morning

-0.31

 Gaunt

-0.31

STRIP

-0.30

POSITIVE LOGITS

tagHelperRunner

0.68

ագրություններ

0.60

EndInit

0.59

TargetException

0.57

elemField

0.55

 informée

0.53

TagMode

0.53

:+:

0.52

rawDesc

0.52

+#+#

0.51

Activations Density 0.081%

terms and concepts related to technical specifications and classifications

semiparametric modeling and technical code snippets

No Comments

No Known Activations