INDEX

Explanations

something that's hard to determine, but it seems to activate on language related to religious faith/practices, surveys, the suffix "ly," and certain proper nouns.

oai_token-act-pair · gemini-2.0-flash

positive sentiment

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_25/width_16k/average_l0_41

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.25.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

 sets

-1.05

 Sets

-0.90

sets

-0.89

Sets

-0.85

 SETS

-0.72

expandindo

-0.66

 useParams

-0.63

Obrázky

-0.63

IntoConstraints

-0.61

 casts

-0.60

POSITIVE LOGITS

Looks

0.69

 Looks

0.65

 looks

0.65

 honours

0.64

 honors

0.62

 benefits

0.58

 flour

0.58

StructEnd

0.56

ilités

0.56

 rewards

0.55

Activations Density 6.640%

something that's hard to determine, but it seems to activate on language related to religious faith/practices, surveys, the suffix "ly," and certain proper nouns.

positive sentiment

No Comments

No Known Activations

something that's hard to determine, but it seems to activate on language related to religious faith/practices, surveys, the suffix "ly," and certain proper nouns.

positive sentiment

No Comments

No Known Activations