INDEX

Explanations

actions and statements regarding attempts, interventions, and criticisms in a formal context

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

 doing

-0.53

 Doing

-0.53

GenerationType

-0.50

 conduct

-0.49

 balance

-0.48

SEGUIR

-0.46

paz

-0.46

 zrobi

-0.45

 gospodar

-0.45

Doing

-0.45

POSITIVE LOGITS

:✨

0.73

contentLoaded

0.67

ContentAlignment

0.65

########.

0.63

+#+#

0.59

 surla

0.56

beiros

0.56

 defaultstate

0.55

 wort

0.54

uxxxx

0.54

Activations Density 0.370%

actions and statements regarding attempts, interventions, and criticisms in a formal context

No Comments

No Known Activations