INDEX

Explanations

affirmations and expressions of agreement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ocu

-0.16

 dÃ¶k

-0.15

Hey

-0.14

acht

-0.14

alian

-0.14

 PERMISSION

-0.13

Uns

-0.13

fet

-0.13

atra

-0.13

mitt

-0.13

POSITIVE LOGITS

 wrong

0.75

 correct

0.75

wrong

0.64

 Wrong

0.60

 WRONG

0.57

Wrong

0.57

correct

0.57

 incorrect

0.56

 Correct

0.53

Correct

0.52

Activations Density 0.226%

affirmations and expressions of agreement

No Comments

No Known Activations