INDEX

Explanations

key phrases and contextual clues about processes or concepts that are under investigation or in discussion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

titleMargin

-0.47

InsertCommand

-0.46

VERTISING

-0.44

gridx

-0.43

ABAD

-0.42

Iné

-0.42

 überlegen

-0.41

 اض

-0.40

řevě

-0.39

 uova

-0.39

POSITIVE LOGITS

 constitutes

0.70

 merupakan

0.69

 đây

0.67

 represents

0.65

Clik

0.64

pinMode

0.62

 serves

0.60

 bezeichneter

0.60

 stanowi

0.59

seamnă

0.59

Activations Density 0.580%

key phrases and contextual clues about processes or concepts that are under investigation or in discussion

No Comments

No Known Activations