INDEX

Explanations

phrases indicating emphasis on self-worth or assertiveness

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

emphatic phrases

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

ագրություններ

-0.64

twimg

-0.60

parsedMessage

-0.59

 שוליים

-0.58

onOptions

-0.56

enumi

-0.56

Erreferentziak

-0.52

UnusedPrivate

-0.50

IActionResult

-0.50

SBATCH

-0.50

POSITIVE LOGITS

хьтан

0.49

 freakin

0.45

 freaking

0.45

windowFixed

0.43

Lorem

0.42

lak

0.41

 fuckin

0.40

 fucking

0.40

 shit

0.39

ItemClick

0.38

Activations Density 0.027%

phrases indicating emphasis on self-worth or assertiveness

emphatic phrases

No Comments

No Known Activations