INDEX

Explanations

references to exposure and contamination

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

mez

-0.15

igmat

-0.15

indle

-0.14

ÐĽÐŀ

-0.14

otos

-0.13

ÑĢÐ°Ð´

-0.13

Calibri

-0.13

FileType

-0.13

elo

-0.12

usat

-0.12

POSITIVE LOGITS

 Natural

0.28

Natural

0.23

jur

0.19

 sourced

0.18

 natural

0.18

 claimed

0.17

-State

0.17

 State

0.17

 Physical

0.17

NAT

0.17

Activations Density 0.000%

references to exposure and contamination

No Comments

No Known Activations