INDEX

Explanations

comparative phrases indicating differences in rates and measurements

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

comparison and proportion

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

bildeten

-0.39

↹

-0.39

 transmits

-0.36

 mentally

-0.36

 Outdoor

-0.36

 Mental

-0.36

 defeating

-0.36

 Sugar

-0.35

 donating

-0.35

 Anhäng

-0.35

POSITIVE LOGITS

RegressionTest

0.56

 betweenstory

0.53

migrationBuilder

0.53

AnchorStyles

0.52

 ligiloj

0.49

tonsoft

0.48

rzost

0.45

RTCK

0.45

copg

0.44

lorette

0.43

Activations Density 0.148%

comparative phrases indicating differences in rates and measurements

comparison and proportion

No Comments

No Known Activations