INDEX

Explanations

expressions of love and emotional expressions

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

"I" followed by "love"

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

 CreateTagHelper

-0.79

 snippetHide

-0.77

LookAnd

-0.77

tagHelperRunner

-0.75

 nahilalakip

-0.74

sizeCache

-0.74

 HttpNotFound

-0.73

 ProtoMessage

-0.73

parsedMessage

-0.72

httphttps

-0.68

POSITIVE LOGITS

 Kommission

0.36

 wouldn

0.34

Thank

0.28

So

0.28

 thank

0.28

 would

0.27

 médecine

0.26

 شك

0.25

 koń

0.25

ith

0.25

Activations Density 0.314%

expressions of love and emotional expressions

"I" followed by "love"

No Comments

No Known Activations