INDEX

Explanations

phrases that encourage the reader to take action or explore further

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

.executeQuery

-0.16

Marker

-0.16

elerik

-0.15

eka

-0.15

xes

-0.15

ului

-0.14

airy

-0.14

odo

-0.14

Ã£

-0.13

iferay

-0.13

POSITIVE LOGITS

out

0.21

ered

0.20

lists

0.18

 ìĤ¬íķŃ

0.16

washer

0.16

strup

0.15

list

0.15

 Nunes

0.14

eeper

0.14

Activations Density 0.020%

phrases that encourage the reader to take action or explore further

No Comments

No Known Activations