INDEX

Explanations

phrases urging action or inquiry for information

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

util

-0.14

ilters

-0.14

 ëģ

-0.13

éļĨ

-0.13

avourites

-0.13

IEWS

-0.13

EEK

-0.13

.cx

-0.13

olith

-0.13

ized

-0.13

POSITIVE LOGITS

lay

0.29

ings

0.23

horn

0.22

DOMNode

0.22

out

0.21

 answers

0.20

LAY

0.19

 which

0.18

NavController

0.18

 more

0.18

Activations Density 0.037%

phrases urging action or inquiry for information

No Comments

No Known Activations