INDEX

Explanations

phrases that indicate encouragement or support

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

followed by prepositions

np_acts-logits-general · gemini-2.0-flash

describes positive attributes and outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

<eos>

-0.35

 richtige

-0.26

gnore

-0.25

 olaca

-0.25

bł

-0.24

 next

-0.24

POSITIVE LOGITS

 kasarigan

0.91

 EconPapers

0.87

parsedMessage

0.85



0.84

 pleaſure

0.82

<unused52>

0.80

richTextPanel

0.80

<unused8>

0.80

<unused3>

0.80

[@BOS@]

0.80

Activations Density 9.724%

phrases that indicate encouragement or support

followed by prepositions

describes positive attributes and outcomes

No Comments

No Known Activations