INDEX

Explanations

instances of the word "lead" and its variations in various contexts

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

lead to or guide

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

 carinho

-0.39

 notwithstanding

-0.39

aringan

-0.38

 Barcelone

-0.37

 Wünsche

-0.37

 поми

-0.37

 réservoir

-0.37

 Atlántico

-0.36

 August

-0.36

 Goldberg

-0.35

POSITIVE LOGITS

 lead

1.30

Lead

1.27

 Lead

1.25

lead

1.24

 LEAD

1.16

leads

1.16

 Leads

1.15

Leads

1.12

 leads

1.12

LEAD

1.08

Activations Density 0.020%

instances of the word "lead" and its variations in various contexts

lead to or guide

No Comments

No Known Activations