INDEX

Explanations

words related to a variety of technical topics including cameras, graphics, software development, and physical objects.

oai_token-act-pair · gemini-2.0-flash

phrases indicating specific details or features

np_token-act-pair-logits · gpt-4o-mini

technical actions or files

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 nakalista

-0.84

 itſelf

-0.73

脚注の使い方

-0.68



-0.65

titleMargin

-0.63

 Administrativna

-0.63

 ddelweddau

-0.62

 betweenstory

-0.60

twimg

-0.59

Portale

-0.58

POSITIVE LOGITS

 with

0.54

 that

0.52

 featuring

0.51

 resourceCulture

0.49

findpost

0.48

 consisting

0.47

 they

0.47

 created

0.45

 objeto

0.45

Activations Density 9.641%

words related to a variety of technical topics including cameras, graphics, software development, and physical objects.

phrases indicating specific details or features

technical actions or files

No Known Activations