INDEX

Explanations

the word stem "do" in various forms and contexts, not necessarily related

oai_token-act-pair · gemini-2.0-flash

done

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_0/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

-0.96

-0.91

do

-0.85

or

-0.84

-0.82

-0.81

in

-0.79

<eos>

-0.78

POSITIVE LOGITS

 Theſe

1.61

 ainfi

1.61

 itſelf

1.57

Efq

1.53

 ſtate

1.48

 ſeveral

1.47

 auffi

1.46

 myſelf

1.46

 becauſe

1.43

 ſmall

1.41

Activations Density 1.781%

the word stem "do" in various forms and contexts, not necessarily related

done

No Comments

No Known Activations