Circuit Sparsity

No connected neurons data available

Explanations

socket.bind or break

python-code · gemini-2.5-flash-lite

block of code after a colon (negative activations)

python-code · gemini-2.5-flash-lite

Auto-Interp Tools

Configuration

Prompts (Dashboard)

16,384 prompts, 256 tokens each

Dataset (Dashboard)

neuronpedia/python-code-simplified

Embeds

IFrame

Link

Negative Logits

 person

-0.11

 student

-0.10

 instruction

-0.09

dp

-0.09

 instructions

-0.08

 neighbors

-0.07

 stack

-0.07

 possible

-0.07

 best

-0.06

 team

-0.06

Positive Logits

graph

0.08

ass

0.08

 pass

0.07

opulation

0.07

_g

0.06

div

0.05

.g

0.05

º

0.05

gh

0.05

Activations Density 100.000%

No Known Activations

Prev Next

Circuit Sparsity

No connected neurons data available

Explanations

opening parenthesis of function calls or definitions (negative activations)

python-code · gemini-2.5-flash-lite

Auto-Interp Tools

Configuration

Prompts (Dashboard)

16,384 prompts, 256 tokens each

Dataset (Dashboard)

neuronpedia/python-code-simplified

Embeds

IFrame

Link

Negative Logits

 break

-0.03

in

-0.03

>=

-0.02

<=

-0.02

[-

-0.02

!=

-0.02

}")↵↵

-0.02

ns

-0.02

}")<|

-0.02

Positive Logits

[j

0.08

instance

0.03

 given

0.03

0.02

log

0.02

ram

0.01

ro

0.01

ases

0.01

Activations Density 89.432%