INDEX

Explanations

a wide variety of somewhat random words that are often parts of other words.

oai_token-act-pair · gemini-2.0-flash

code/shopping

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

Not in Any Lists

No Comments

Negative Logits

contentLoaded

-0.93

 CreateTagHelper

-0.81

findpost

-0.74

RenderAtEndOf

-0.73

AddTagHelper

-0.71

 resourceCulture

-0.71

Rptr

-0.69

AddHtmlAttribute

-0.69

 AssemblyCulture

-0.68

 fallu

-0.66

POSITIVE LOGITS

iostream

0.53

 ngoài

0.50

chur

0.49

pen

0.49

toli

0.47

pur

0.47

tolo

0.46

NOC

0.46

いかがでしょうか

0.45

tése

0.45

Activations Density 10.724%

a wide variety of somewhat random words that are often parts of other words.

code/shopping

No Comments

No Known Activations