INDEX

Explanations

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.

oai_token-act-pair · o4-mini Triggered by @jyhe0408

checking number of arguments or length

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_22/width_131k

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 grå

-1.09

Déf

-1.06

 水晶

-1.05

Specifik

-1.02

 wohnung

-1.01

 bisous

-1.00

恐れ

-1.00

なども

-0.99

 定制

-0.97

 merve

-0.96

POSITIVE LOGITS

than

1.15

1.11

one

1.10

1.05

1.02

})

1.02

figsize

1.02

0.97

Ag

0.97

ans

0.97

Activations Density 0.001%

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.

checking number of arguments or length

No Comments

No Known Activations

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.

checking number of arguments or length

No Comments

No Known Activations

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ­ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.

checking number of arguments or length

No Comments

No Known Activations

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ­ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.

checking number of arguments or length

No Comments

No Known Activations

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.

The neuron fires on short code tokens made up of symbols (and symbolic “words”) such as comparison operators (==, >, ne), command-line flags (‐p), or API flag identifiers (VAR in METH_VARARGS)—in other words, it detects symbolic/operator tokens in code.