INDEX

Explanations

This neuron activates on a variety of seemingly unrelated words and characters, possibly indicating a failure condition or sensitivity to a very specific and uncommon pattern.

oai_token-act-pair · gemini-2.0-flash

Technical/Scientific Writing

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_12/width_16k/average_l0_6

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 froide

-0.40

 antigüedad

-0.40

HomeComponent

-0.40

ל

-0.39

 químicas

-0.37

 convivencia

-0.37

 correctly

-0.37

hæng

-0.36

 complètes

-0.35

 Amber

-0.35

POSITIVE LOGITS

 AssemblyCulture

1.07

 समीक्षक

1.05

 незавершена

1.01

ValueStyle

0.99

AutoScaleMode

0.98

DockStyle

0.93

DebuggerStep

0.91

 ModelExpression

0.88

енча

0.88

出版年

0.87

Activations Density 1.071%

This neuron activates on a variety of seemingly unrelated words and characters, possibly indicating a failure condition or sensitivity to a very specific and uncommon pattern.

Technical/Scientific Writing

No Comments

No Known Activations

This neuron activates on a variety of seemingly unrelated words and characters, possibly indicating a failure condition or sensitivity to a very specific and uncommon pattern.

Technical/Scientific Writing

No Comments

No Known Activations