INDEX

Explanations

references to emotional pain and suffering

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

atisfaction

-0.15

linger

-0.15

peria

-0.15

ulong

-0.15

 luder

-0.14

 weiber

-0.14

itelist

-0.14

ìŀĳ

-0.14

ongan

-0.13

MUX

-0.13

POSITIVE LOGITS

har

0.28

 hell

0.27

 darkest

0.23

 grim

0.23

 dark

0.22

som

0.22

 pain

0.21

Hell

0.21

 distress

0.21

 difficulty

0.20

Activations Density 0.027%

references to emotional pain and suffering

No Comments

No Known Activations