INDEX

Explanations

instances of emotional reactions and social observations

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

its

-0.20

here

-0.16

nt

-0.15

ade

-0.15

....

-0.15

ve

-0.14

 bazen

-0.14

....↵

-0.14

ge

-0.14

POSITIVE LOGITS

 period

0.18

 haha

0.17

 PLUS

0.17

sans

0.17

 Ãł

0.17

er

0.17

lol

0.16

LOL

0.16

ha

0.16

eh

0.16

Activations Density 0.224%

instances of emotional reactions and social observations

No Comments

No Known Activations