INDEX

Explanations

phrases associated with influential statements or assertions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

followed by "one"

np_acts-logits-general · gemini-2.0-flash

this one

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

templateUrl

-0.46

Set

-0.44

standig

-0.41

Set

-0.41

Przypisy

-0.41

-0.40

FieldDescriptor

-0.40

gar

-0.39

POSITIVE LOGITS

 ones

2.32

 celui

1.51

 quelli

1.46

 Ones

1.39

 yours

1.39

 quello

1.36

 ours

1.36

 theirs

1.36

 Celui

1.30

 hers

1.27

Activations Density 0.649%

phrases associated with influential statements or assertions

followed by "one"

this one

No Comments

No Known Activations