INDEX

Explanations

affirmative and positive expressions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.Messaging

-0.15

REA

-0.14

 lands

-0.14

acin

-0.13

prim

-0.13

icha

-0.13

ãĤĪãģĨãģ§ãģĻ

-0.13

çĶ

-0.13

reff

-0.13

zin

-0.13

POSITIVE LOGITS

 relationships

0.46

 relationship

0.45

 Relationships

0.39

 interaction

0.38

 interactions

0.37

 relations

0.37

relationship

0.37

 bond

0.36

 Relationship

0.36

relationships

0.35

Activations Density 0.020%

affirmative and positive expressions

No Comments

No Known Activations