INDEX

Explanations

phrases that convey caution or warnings about certain situations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_22/width_16k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.hook_mlp_out

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Datuak

-0.57

FunctionFlags

-0.57

umbang

-0.52

Autoritní

-0.52

 häm

-0.50

 rallado

-0.50

Супер

-0.50

 Estimates

-0.50

 CreateTagHelper

-0.49

{?>

-0.49

POSITIVE LOGITS

 avoid

1.75

 avoided

1.71

 avoiding

1.67

 avoids

1.62

avoid

1.55

 Avoid

1.51

 avoidance

1.48

Avoid

1.46

 Avoiding

1.37

 AVOID

1.28

Activations Density 0.399%

phrases that convey caution or warnings about certain situations

No Comments

No Known Activations