INDEX

Explanations

complex mathematical expressions and symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

linger

-0.17

olean

-0.17

oref

-0.15

inea

-0.15

digital

-0.15

ono

-0.15

ject

-0.15

 Fore

-0.15

argon

-0.14

harma

-0.14

POSITIVE LOGITS

le

0.37

ge

0.34

ne

0.28

gne

0.27

ge

0.22

nge

0.22

gg

0.21

ll

0.20

in

0.18

agt

0.18

Activations Density 0.071%

complex mathematical expressions and symbols

No Comments

No Known Activations