INDEX

Explanations

genome sequence

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

terms related to genomics and genetic sequencing, especially the word "genome" and related scientific terminology.

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

The neuron activates on occurrences of “genome” and its close variants (e.g. genomic, genomes, sequence) in the text.

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

 That

-2.73

</i>

-2.70

-2.67

𞤢

-2.53

 They

-2.23

...");

-2.14

 This

-2.02

⡑

-2.00

焄

-2.00

 αυτή

-1.98

POSITIVE LOGITS

of

2.81

 succès

2.09

нё

2.02

阂

1.98

 балко

1.88

 جديدة

1.88

ının

1.80

 trivi

1.77

再来

1.77

 untuk

1.77

Activations Density 0.020%

genome sequence

terms related to genomics and genetic sequencing, especially the word "genome" and related scientific terminology.

The neuron activates on occurrences of “genome” and its close variants (e.g. genomic, genomes, sequence) in the text.

No Comments

No Known Activations

genome sequence

terms related to genomics and genetic sequencing, especially the word "genome" and related scientific terminology.

The neuron activates on occurrences of “genome” and its close variants (e.g. genomic, genomes, sequence) in the text.

No Comments

No Known Activations