INDEX

Explanations

instances of the word "moderate" and its variations related to intensity or degree

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

moderate intensity or dose

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

 FileNotFound

-0.53

 pellic

-0.52

 eyelash

-0.52

ckså

-0.51

 inkjet

-0.49

 pelota

-0.46

Ellips

-0.46

getJob

-0.45

Gol

-0.45

getKey

-0.45

POSITIVE LOGITS

 Moderate

0.96

 moderate

0.90

moderate

0.83

 moderated

0.82

Moderate

0.79

 Moder

0.73

 moderately

0.72

 moderation

0.71

moder

0.70

MODER

0.68

Activations Density 0.006%

instances of the word "moderate" and its variations related to intensity or degree

moderate intensity or dose

No Comments

No Known Activations