INDEX

Explanations

phrases that describe methods or approaches to achieve something

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

best way to do

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

 ContentValues

-0.68

ſelves

-0.66

 ModelExpression

-0.65

сылкі

-0.64

ſelf

-0.63

intios

-0.63

ðsíða

-0.62

tagHelperRunner

-0.62

dataclass

-0.61

 juſ

-0.60

POSITIVE LOGITS

 best

0.64

best

0.57

 najlepiej

0.52

 conviene

0.45

 besten

0.45

 terbaik

0.44

 sebaiknya

0.44

 بهترین

0.43

Best

0.43

 лучше

0.42

Activations Density 0.023%

phrases that describe methods or approaches to achieve something

best way to do

No Comments

No Known Activations