INDEX

Explanations

a variety of proper names and titles associated with individuals and roles in a community or institution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

Ã¶y

-0.15

ubby

-0.15

è©±

-0.15

ÄĽj

-0.15

imenti

-0.15

stown

-0.15

ead

-0.14

 stub

-0.14

ordable

-0.14

cape

-0.14

POSITIVE LOGITS

wal

0.23

odia

0.21

olia

0.17

izada

0.16

oria

0.16

ÑħÑĥ

0.15

Roy

0.15

urve

0.15

hani

0.15

aria

0.15

Activations Density 0.129%

a variety of proper names and titles associated with individuals and roles in a community or institution

No Comments

No Known Activations