INDEX

Explanations

specific names or terms related to scientific datatypes and statistical methods

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

words starting with l

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 noDo

-0.50

 CURIAM

-0.47

beiro

-0.45

 Guten

-0.45

tableFuture

-0.44

 Савезне

-0.43

 UnityEditor

-0.43

 cannibal

-0.42

 jButton

-0.42

 numberWith

-0.41

POSITIVE LOGITS

Ljava

0.63

lorn

0.61

LK

0.61

LK

0.59

LF

0.59

 Liza

0.57

lie

0.57

luster

0.57

 Lilly

0.57

 Linton

0.57

Activations Density 1.723%

specific names or terms related to scientific datatypes and statistical methods

words starting with l

No Comments

No Known Activations