INDEX

Explanations

This neuron seems to be activating for a variety of words, with some patterns including proper nouns, creative writing, and possibly science fiction.

oai_token-act-pair · gemini-2.0-flash

fictional characters and stories

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_4/width_16k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 itſelf

-0.90

 myſelf

-0.89

 pleaſure

-0.83

LookAnd

-0.81

 neceff

-0.80

 kaarangay

-0.80

 Monfieur

-0.80

Билгалдахарш

-0.79

 purpoſe

-0.79

 againſt

-0.78

POSITIVE LOGITS

 CreateTagHelper

0.49

0.47

 concernant

0.47

),

0.46

0.45

 fordi

0.45

 knji

0.42

0.41

Activations Density 2.047%

This neuron seems to be activating for a variety of words, with some patterns including proper nouns, creative writing, and possibly science fiction.

fictional characters and stories

No Known Activations