INDEX

Explanations

relationships and adaptations in models related to directed graphical structures

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

sentence endings before new points

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

 المعيارى

-0.63

 queſta

-0.57

RTEX

-0.57

Portail

-0.56

tvguidetime

-0.52

SharedDtor

-0.51

 navideño

-0.51

 feux

-0.49

 OMITBAD

-0.49

eschön

-0.48

POSITIVE LOGITS

UserScript

0.45

 estekak

0.39

ynb

0.38

 jsPsych

0.35

AutoresizingMask

0.35

+#+#

0.33

DllImport

0.33

 сход

0.33

::::::::

0.32

onAttach

0.32

Activations Density 0.083%

relationships and adaptations in models related to directed graphical structures

sentence endings before new points

No Comments

No Known Activations