INDEX

Explanations

references to archaeological discoveries, particularly ancient human artifacts or fossils

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

angu

-0.15

 ÎĶÎ·

-0.13

estre

-0.13

serter

-0.13

 stalking

-0.12

aggio

-0.12

edian

-0.12

æłª

-0.12

.esp

-0.12

;element

-0.12

POSITIVE LOGITS

 archae

0.41

 archaeological

0.38

 Archae

0.36

 excav

0.36

arch

0.36

 digs

0.36

ar

0.34

 excavation

0.33

dig

0.33

-exc

0.32

Activations Density 0.272%

references to archaeological discoveries, particularly ancient human artifacts or fossils

No Comments

No Known Activations