INDEX

Explanations

common phrases or structures in language that pertain to possession or association

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

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

-0.54

定的

-0.54

Be

-0.53

id

-0.53

NY

-0.52

異なる

-0.52

zu

-0.52

 uniquely

-0.52

POSITIVE LOGITS

ContentAsync

1.07

 ExecuteAsync

1.05

 Мексичка

1.04

 Roskov

0.94

 utafitiHapana

0.93

 ProtoMessage

0.91

 виправивши

0.90

 meste

0.89

 vět

0.88

 BoxFit

0.88

Activations Density 0.102%

common phrases or structures in language that pertain to possession or association

No Comments

No Known Activations