INDEX

Explanations

references to popular media and cultural institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ugin

-0.17

oub

-0.16

stab

-0.16

jury

-0.15

erdem

-0.15

ëĵľë¦¬

-0.15

reur

-0.14

beer

-0.14

POSITIVE LOGITS

rames

0.16

chem

0.15

é¼

0.14

antee

0.14

/platform

0.13

çĹĩ

0.13

#,

0.13

 LENG

0.13

!,

0.13

arm

0.13

Activations Density 0.063%

references to popular media and cultural institutions

No Comments

No Known Activations