INDEX

Explanations

phrases indicating generalizations or commonly observed patterns

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

generally followed by a verb

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

SBATCH

-0.58

PathVariable

-0.57

inghouse

-0.55

 inactivation

-0.54

poptotic

-0.54

ituaries

-0.54

جوايز

-0.54

spyOn

-0.54

 shavings

-0.53

 resistivity

-0.52

POSITIVE LOGITS

 generally

2.03

generally

1.82

 Generally

1.71

Generally

1.67

 generalmente

1.36

 généralement

1.22

 umumnya

1.20

 geralmente

1.20

 typically

1.06

 usually

1.03

Activations Density 0.011%

phrases indicating generalizations or commonly observed patterns

generally followed by a verb

No Comments

No Known Activations