INDEX

Explanations

phrases emphasizing the concept of being the "best" or "greatest" in some context

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

superlatives

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

Negative Logits



-0.57

Попис

-0.53

)')

-0.51

ربای

-0.51

 Chwiliwch

-0.50

OGND

-0.50

voix

-0.50

ingle

-0.48

dafx

-0.47

WithIOException

-0.47

POSITIVE LOGITS

printStackTrace

0.60

 nakalista

0.46

 IConfiguration

0.43

 محفوظة

0.42

 ſon

0.40

Tym

0.40

WRONG

0.40

 siga

0.39

 conquista

0.39

 pacto

0.38

Activations Density 0.004%

phrases emphasizing the concept of being the "best" or "greatest" in some context

superlatives

No Known Activations