INDEX

Explanations

the neuron is finding words related to research and data collection

oai_token-act-pair · gemini-2.0-flash

words related to recording or documenting

np_token-act-pair-logits · gpt-4o-mini

recording

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

IsContent

-0.67

elemField

-0.64

 ujednoznacz

-0.61

 surla

-0.60

enderror

-0.59

DispatchToProps

-0.57

PageRoute

-0.55

KURZBESCHREIBUNG

-0.54

قلاب

-0.51

 polenta

-0.49

POSITIVE LOGITS

 recording

1.93

 recorded

1.73

 record

1.64

recording

1.63

Recording

1.59

 Recording

1.56

 documenting

1.56

recorded

1.52

record

1.43

 recorder

1.43

Activations Density 4.385%

the neuron is finding words related to research and data collection

words related to recording or documenting

recording

No Comments

No Known Activations