INDEX

Explanations

the keywords 'for' and 'in', used for looping in programming, as well as the keyword 'as'

oai_token-act-pair · gemini-2.0-flash

programming control flow keywords

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

for

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

Not in Any Lists

No Comments

Negative Logits

 itſelf

-1.11

 myſelf

-0.85

 Theſe

-0.80

 becauſe

-0.78

 Inſ

-0.78

 ſever

-0.77

 Diſ

-0.76

 himſelf

-0.76

 pleaſure

-0.75

 ་་

-0.74

POSITIVE LOGITS

to

0.75

0.67

0.64

0.63

can

0.62

0.61

gemeen

0.60

 move

0.60

0.59

Activations Density 0.661%

the keywords 'for' and 'in', used for looping in programming, as well as the keyword 'as'

programming control flow keywords

for

No Comments

No Known Activations