INDEX

Explanations

terms related to television networks and governance structures

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

Big Three affiliates

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

المكان

-0.41

ProcessEvent

-0.40

PreInfinity

-0.39

aarrggbb

-0.37

databind

-0.35

 Davey

-0.33

yyj

-0.33

Advisor

-0.33

 repair

-0.32

 Window

-0.32

POSITIVE LOGITS

bootstrapcdn

0.54

tinyos

0.49

Chham

0.47

 createSlice

0.46

 hermanas

0.46

 Trilogy

0.43

AddTagHelper

0.43

rungsseite

0.42

 trio

0.42

 közül

0.42

Activations Density 0.113%

terms related to television networks and governance structures

Big Three affiliates

No Comments

No Known Activations