INDEX

Explanations

loss function

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

the word "loss" in technical or scientific contexts, particularly in machine learning, signal processing, and medical/scientific literature.

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

The neuron fires on occurrences of the word “loss” (including technical phrases like “custom loss,” “loss function,” “triplet loss,” etc.).

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

-2.53

-2.30

喁

-2.25

 victimes

-1.96

 analisis

-1.95

以上です

-1.95

鲔

-1.94

玧

-1.91

 этого

-1.88

ｄ

-1.88

POSITIVE LOGITS

3.05

2.36

2.27

2.13

2.08

2.06

也

1.95

1.89

也

1.88

1.87

Activations Density 0.002%

loss function

the word "loss" in technical or scientific contexts, particularly in machine learning, signal processing, and medical/scientific literature.

The neuron fires on occurrences of the word “loss” (including technical phrases like “custom loss,” “loss function,” “triplet loss,” etc.).

No Comments

No Known Activations

loss function

the word "loss" in technical or scientific contexts, particularly in machine learning, signal processing, and medical/scientific literature.

The neuron fires on occurrences of the word “loss” (including technical phrases like “custom loss,” “loss function,” “triplet loss,” etc.).

No Comments

No Known Activations