INDEX

Explanations

addresses and location information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

osemite

-0.17

kate

-0.15

meg

-0.15

ascript

-0.14

Hus

-0.14

ivr

-0.14

opping

-0.14

ipt

-0.14

ibir

-0.14

POSITIVE LOGITS

Quad

0.18

 Quad

0.17

Cow

0.17

 quad

0.16

Cow

0.16

Cad

0.15

.portal

0.15

å¥Ī

0.15

 Duncan

0.15

 scanner

0.15

Activations Density 0.013%

addresses and location information

No Comments

No Known Activations