INDEX

Explanations

updates and changes in a process or system

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

finding solutions or updates

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

 Italij

-0.41

ovým

-0.40

 peluang

-0.40

 péché

-0.38

 abstrato

-0.38

 risques

-0.37

 edades

-0.35

 cuánt

-0.35

 exempel

-0.34

 håll

-0.34

POSITIVE LOGITS

httphttps

0.61

queryInterface

0.59

setVerticalGroup

0.56

ConstraintMaker

0.55

клопе

0.55

AnchorStyles

0.54

Diwedd

0.53

expandindo

0.52

 aDecoder

0.52

migrationBuilder

0.50

Activations Density 0.024%

updates and changes in a process or system

finding solutions or updates

No Comments

No Known Activations