INDEX

Explanations

phrases and terms related to "direct" in various contexts

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

direct followed by noun

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 newName

-0.69

ChildScrollView

-0.64

덟

-0.59

 oscillators

-0.56

 classNames

-0.55

 endTime

-0.54

 appName

-0.54

 meisje

-0.54

 eventName

-0.54

 playlists

-0.53

POSITIVE LOGITS

direct

2.17

 direct

2.16

Direct

2.08

 Direct

2.02

 DIRECT

1.73

DIRECT

1.70

 directe

1.51

 directa

1.50

irect

1.35

 directos

1.34

Activations Density 0.020%

phrases and terms related to "direct" in various contexts

direct followed by noun

No Comments

No Known Activations