INDEX

Explanations

phrases indicating a desire to learn about various subjects or topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ä¹ĭä¸Ģ

-0.16

omal

-0.15

gel

-0.15

Cle

-0.14

 figure

-0.14

pel

-0.14

 mast

-0.14

CKET

-0.13

á»Ł

-0.13

 convergence

-0.13

POSITIVE LOGITS

hue

0.15

ekl

0.15

Ø±Ùģ

0.14

.jdesktop

0.14

ABCDEFGHIJKLMNOPQRSTUVWXYZ

0.14

íĬ¹ë³Ħ

0.14

legg

0.13

iard

0.13

Ð³Ð¾ÑĢÐ¾Ð´

0.13

 eskort

0.13

Activations Density 0.020%

phrases indicating a desire to learn about various subjects or topics

No Comments

No Known Activations