INDEX

Explanations

phrases related to variations and specific metrics in research contexts

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

other, others

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

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

pantalones

-0.44

enumi

-0.37

 Schmel

-0.36

")"

-0.35

ůvod

-0.35

AsUp

-0.35

richTextPanel

-0.35

cabulary

-0.35

"}"

-0.34

 Brunnen

-0.34

POSITIVE LOGITS

 other

0.52

 otherwise

0.48

autres

0.47

 getattr

0.46

fjspx

0.46

onActivityResult

0.46

 versions

0.45

 others

0.45

 остальные

0.45

 PeEnEo

0.44

Activations Density 0.101%

phrases related to variations and specific metrics in research contexts

other, others

No Comments

No Known Activations