INDEX

Explanations

overthrow and connectionString

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

the word "overthrow" or code/technical terms like "setInt" and "connectionString" in programming contexts.

oai_token-act-pair · claude-4-5-sonnet Triggered by @bfgcn7frf5

This neuron responds to code-style identifiers and programming keywords (e.g. method names like “setInt,” property names like “connectionString,” and keywords/substrings like “throw” or version tokens).

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

New Auto-Interp

Configuration

google/gemma-scope-27b-pt-res/layer_10/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

-2.91

𔘓

-2.56

 marvel

-2.52

↵↵

-2.28

was

-2.27

 così

-2.19

''

-2.16

这里

-2.14

 nuova

-2.11

纐

-2.09

POSITIVE LOGITS

籐

2.63

zembro

2.39

 aña

2.38



2.27

};*/

2.25



2.23

).

2.17

轳

2.17

ਞ

2.14

…).

2.13

Activations Density 0.001%

overthrow and connectionString

the word "overthrow" or code/technical terms like "setInt" and "connectionString" in programming contexts.

This neuron responds to code-style identifiers and programming keywords (e.g. method names like “setInt,” property names like “connectionString,” and keywords/substrings like “throw” or version tokens).

No Comments

No Known Activations

overthrow and connectionString

the word "overthrow" or code/technical terms like "setInt" and "connectionString" in programming contexts.

This neuron responds to code-style identifiers and programming keywords (e.g. method names like “setInt,” property names like “connectionString,” and keywords/substrings like “throw” or version tokens).

No Comments

No Known Activations