INDEX

Explanations

phrases indicating dependency or absence

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

"Without" followed by a word

np_acts-logits-general · gemini-2.0-flash

without doubt

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 Cæsar

-0.91

 Monfieur

-0.82

 houſe

-0.81

Efq

-0.77

 pleaſure

-0.77

 Jefus

-0.73

ſelf

-0.72

 againſt

-0.72

 ſever

-0.70

 AssemblyCulture

-0.68

POSITIVE LOGITS

any

0.94

 Without

0.93

Ohne

0.89

Without

0.87

 having

0.85

 without

0.82

 knowing

0.81

0.80

without

0.80

 WITHOUT

0.77

Activations Density 0.077%

phrases indicating dependency or absence

"Without" followed by a word

without doubt

No Comments

No Known Activations