INDEX

Explanations

specific names and details related to museums and their collections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

upo

-0.13

anyl

-0.13

.statusText

-0.13

 Peaks

-0.12

 Activate

-0.12

 Pussy

-0.12

kÃ¡ch

-0.12

ç«ľ

-0.12

avana

-0.12

 onCreateView

-0.12

POSITIVE LOGITS

 museum

0.50

 Museum

0.50

 Muse

0.45

 museums

0.44

 collections

0.42

 muse

0.41

mus

0.40

 Collections

0.38

Mus

0.38

åįļçī©

0.37

Activations Density 0.203%

specific names and details related to museums and their collections

No Comments

No Known Activations