INDEX

Explanations

relevant prepositions and conjunctions that indicate relationships between elements in a sentence

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

times and durations

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

the neuron detects numeric tokens (numbers, years, times, and other digit-containing tokens).

oai_token-act-pair · gpt-5-mini Triggered by @vetterc0

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 Infórmanos

-0.63

Jeografia

-0.62

oredCriteria

-0.55

 surla

-0.54

chartInstance

-0.53

########.

-0.52

oa̍t

-0.52

pexpr

-0.52

ArrowToggle

-0.51

AsUp

-0.51

POSITIVE LOGITS

GIF

0.36

 politi

0.35

 مشين

0.35

Formatted

0.35

 dever

0.35

 guten

0.35

gif

0.34

춘

0.34

 NCERT

0.34

<()>

0.34

Activations Density 0.291%

relevant prepositions and conjunctions that indicate relationships between elements in a sentence

times and durations

the neuron detects numeric tokens (numbers, years, times, and other digit-containing tokens).

No Comments

No Known Activations