INDEX

Explanations

actions related to removal or erasure

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

Ends with "ing" or "ed" followed by "out"

np_acts-logits-general · gemini-2.0-flash

verbs followed by out

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

Personendaten

-0.80

 AssemblyCulture

-0.76

IsMutable

-0.71

MessageOf

-0.69

GEBURTSDATUM

-0.66

Portale

-0.66

RegistryLite

-0.63

WaitGroup

-0.63

Geplaatst

-0.62

 ModelRenderer

-0.62

POSITIVE LOGITS

out

1.33

up

1.27

off

0.95

 down

0.93

 forth

0.83

up

0.72

 away

0.72

out

0.69

 back

0.69

 into

0.67

Activations Density 0.310%

actions related to removal or erasure

Ends with "ing" or "ed" followed by "out"

verbs followed by out

No Comments

No Known Activations