INDEX

Explanations

The neuron is picking out tokens inside quoted speech (i.e. direct quotations).

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

first-person statements expressing personal hopes, expectations, or aspirations.

oai_token-act-pair · claude-4-5-sonnet Triggered by @jyhe0408

direct-quote statements featuring first-person language expressing opinions, experiences, or intentions.

oai_token-act-pair · gpt-5 Triggered by @jyhe0408

New Auto-Interp

Configuration

google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium

Prompts (Dashboard)

392,802 prompts, 256 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 "...

1.05

 উক্ত

0.98

Whilst

0.96

த்தினை

0.96

 हेतु

0.93

또한

0.91

 Whilst

0.91

にて

0.88

 whilst

0.86

亦

0.86

POSITIVE LOGITS

1.41

!”

1.38

すごく

1.38

！」

1.36

],”

1.25

,”

1.20

 teníamos

1.19

,’’

1.18

！”

1.16

 અમારા

1.16

Activations Density 0.366%

The neuron is picking out tokens inside quoted speech (i.e. direct quotations).

first-person statements expressing personal hopes, expectations, or aspirations.

direct-quote statements featuring first-person language expressing opinions, experiences, or intentions.

No Comments

No Known Activations

The neuron is picking out tokens inside quoted speech (i.e. direct quotations).

first-person statements expressing personal hopes, expectations, or aspirations.

direct-quote statements featuring first-person language expressing opinions, experiences, or intentions.

No Comments

No Known Activations