INDEX

Explanations

phrases related to belief and persuasion

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 cent

-0.16

ä¹ī

-0.14

 Dana

-0.14

han

-0.14

 move

-0.13

aces

-0.13

_corpus

-0.13

à¸´à¸į

-0.13

uzu

-0.13

 Ð¿Ð¾Ð²ÐµÑĢ

-0.13

POSITIVE LOGITS

erse

0.17

.sap

0.16

 toÃ¡n

0.15

enery

0.15

viso

0.15

ropa

0.14

cone

0.14

iosper

0.14

ãĤıãģĽ

0.14

xad

0.14

Activations Density 0.007%

phrases related to belief and persuasion

No Comments

No Known Activations