INDEX

Explanations

supportive phrases and requests for assistance

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

Instead of, sorry, maybe

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

 iprot

-0.66

astify

-0.55

ChildScrollView

-0.54

Personendaten

-0.53

OGND

-0.52

########.

-0.50

(!__

-0.50

rrggbb

-0.49

tangentMode

-0.49

曖昧さ

-0.49

POSITIVE LOGITS

 зато

0.66

 sorry

0.63

 Instead

0.62

 Sorry

0.61

maybe

0.61

 instead

0.59

 maybe

0.59

Instead

0.58

Sorry

0.57

代わりに

0.57

Activations Density 0.138%

supportive phrases and requests for assistance

Instead of, sorry, maybe

No Comments

No Known Activations