INDEX

Explanations

assertive language and calls to action regarding communication and expressing truths

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

aho

-0.15

arp

-0.15

ħĮ

-0.14

Vie

-0.14

Ø¬Ø¯

-0.13

 ëª°

-0.13

oggled

-0.13

È

-0.13

chet

-0.13

itsu

-0.13

POSITIVE LOGITS

 loud

0.41

 Loud

0.36

loud

0.35

 publicly

0.34

 aloud

0.28

 loudly

0.27

 louder

0.27

 clearly

0.27

LOUD

0.26

 public

0.26

Activations Density 0.621%

assertive language and calls to action regarding communication and expressing truths

No Comments

No Known Activations