INDEX

Explanations

exclamation marks used for emphasis or strong feelings

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

ninger

-0.16

CEE

-0.15

estre

-0.15

 DÆ°Æ¡ng

-0.15

ighton

-0.14

erva

-0.14

iously

-0.14

loat

-0.14

riere

-0.14

POSITIVE LOGITS

[](

0.17

 ãĥ¾

0.16

asin

0.15

braco

0.14

\Active

0.14

Ã¤ter

0.14

abella

0.14

éĥİ

0.14

 Rope

0.14

arena

0.14

Activations Density 0.073%

exclamation marks used for emphasis or strong feelings

No Comments

No Known Activations