INDEX

Explanations

conditional phrases and logical connectors

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

having

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

исленность

-0.58

 snippetHide

-0.57

glBegin

-0.49

olume

-0.49

écnicas

-0.47

hearsal

-0.45

 Arrangement

-0.45

CloseOperation

-0.45

PERATURE

-0.44

 mourut

-0.44

POSITIVE LOGITS

 having

0.94

Having

0.80

having

0.78

 Having

0.75

 memiliki

0.69

 иметь

0.69

拥有

0.68

 Memiliki

0.66

 HAVING

0.65

 داشتن

0.64

Activations Density 0.135%

conditional phrases and logical connectors

having

No Comments

No Known Activations