INDEX

Explanations

references to failure and consequences in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

esen

-0.16

itre

-0.15

ething

-0.15

eing

-0.14

iones

-0.14

-0.13

inea

-0.13

achs

-0.13

undy

-0.13

umer

-0.13

POSITIVE LOGITS

big

0.47

BIG

0.40

big

0.40

BIG

0.33

-big

0.32

å¤§

0.30

Big

0.29

_big

0.27

 spectacular

0.27

Big

0.26

Activations Density 0.259%

references to failure and consequences in various contexts

No Comments

No Known Activations