INDEX

Explanations

requests for assistance or help

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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±a

-0.14

éĶĢ

-0.13

.Month

-0.13

mour

-0.13

iland

-0.13

uen

-0.13

.workflow

-0.13

 éĶ

-0.13

ascus

-0.12

ADDE

-0.12

POSITIVE LOGITS

 marshal

0.27

 marsh

0.26

bib

0.26

 race

0.25

 scrut

0.24

aid

0.24

 Marsh

0.24

 Race

0.23

Marshal

0.22

 briefing

0.22

Activations Density 0.026%

requests for assistance or help

No Comments

No Known Activations