INDEX

Explanations

induce apoptosis for gene

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

verbs that indicate something contains, demonstrates, or explains a concept.

oai_token-act-pair · claude-3-7-sonnet-20250219 Triggered by @neilrathi

The neuron is keyed to numeric tokens and numerical expressions (e.g. counts, measurements, percentages, decimal values).

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

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

and

-1.47

jsoup

-1.20

 zowel

-1.17

yeezy

-1.16

了一些

-1.15

for

-1.09

 reveló

-1.09

 主要

-1.07

 diferentes

-1.07

 начинает

-1.06

POSITIVE LOGITS

1.48

ditionally

1.42

strictEqual

1.31

Furthermore

1.28

1.26

Nevertheless

1.24

Meanwhile

1.21

})();

1.20

providedIn

1.17

UNIDENTIFIED

1.16

Activations Density 0.080%

induce apoptosis for gene

verbs that indicate something contains, demonstrates, or explains a concept.

The neuron is keyed to numeric tokens and numerical expressions (e.g. counts, measurements, percentages, decimal values).

No Comments

No Known Activations

induce apoptosis for gene

verbs that indicate something contains, demonstrates, or explains a concept.

The neuron is keyed to numeric tokens and numerical expressions (e.g. counts, measurements, percentages, decimal values).

No Comments

No Known Activations