INDEX

Explanations

specific numerical data and statistical results

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

percentages and measurements

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

 Chwiliwch

-0.63

antMatchers

-0.62

 fasc

-0.57

MessageState

-0.57

sizeCache

-0.57

TypedDataSet

-0.56

qtype

-0.56

évaluateur

-0.56

InitStruct

-0.56

ArgsConstructor

-0.56

POSITIVE LOGITS

 ones

0.36

 yalnızca

0.36

 verdadeira

0.34

 siquiera

0.33

 própria

0.33

 kerugian

0.32

 beira

0.30

 elsewhere

0.30

 lembran

0.30

 sekal

0.29

Activations Density 0.059%

specific numerical data and statistical results

percentages and measurements

No Comments

No Known Activations