INDEX

Explanations

references to the word "sal" and its variations, indicating a focus on related terms or names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

edit

-0.17

edic

-0.16

eding

-0.15

bare

-0.15

èĵ

-0.14

.="

-0.14

aro

-0.14

list

-0.14

elay

-0.13

ãĤĵãģ©

-0.13

POSITIVE LOGITS

zburg

0.23

inity

0.23

utations

0.21

sal

0.20

utation

0.20

omon

0.20

erno

0.20

Sal

0.19

ient

0.19

ience

0.18

Activations Density 0.010%

references to the word "sal" and its variations, indicating a focus on related terms or names

No Comments

No Known Activations