INDEX

Explanations

punctuations and fillers in conversation or narrative

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.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

 carriera

-0.57

CompleteListener

-0.55

 Kond

-0.51

 careers

-0.51

 owner

-0.51

 career

-0.50

 chod

-0.49

TemporalType

-0.49

idyl

-0.48

ArgsConstructor

-0.47

POSITIVE LOGITS

Suomen

0.65

MessageTagHelper

0.63

 linkovi

0.58

 OGSÅ

0.57

setVerticalGroup

0.54

vitis

0.53

Noter

0.53

 تانيه

0.53

imarães

0.52

kháu

0.51

Activations Density 0.491%

punctuations and fillers in conversation or narrative

No Comments

No Known Activations