INDEX

Explanations

expressions related to personal introspection and effort

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

lably

-0.54

plet

-0.52

tiftung

-0.49

 Tratt

-0.49

msg

-0.48

лектри

-0.47

rdom

-0.47

 Schengen

-0.47

vores

-0.47

gio

-0.47

POSITIVE LOGITS

Hochspringen

0.67

 doInBackground

0.66

Portail

0.66

 Baillargeon

0.66

 виправивши

0.65

Aiheesta

0.64

 protoimpl

0.63

EndInit

0.63

SerializedSize

0.62

Autoritní

0.62

Activations Density 0.238%

expressions related to personal introspection and effort

No Comments

No Known Activations