INDEX

Explanations

specific nouns or proper names, particularly those associated with individuals or brands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

combe

-0.17

elik

-0.16

onne

-0.16

mate

-0.14

.NewRequest

-0.14

utations

-0.14

AMAGE

-0.14

Ð½Ð¸ÑĨÐ¸Ð¿

-0.14

ivot

-0.14

assy

-0.14

POSITIVE LOGITS

cro

0.19

Cro

0.18

Cro

0.17

asco

0.16

gili

0.16

adil

0.16

gro

0.15

Gro

0.15

gro

0.15

zam

0.14

Activations Density 0.032%

specific nouns or proper names, particularly those associated with individuals or brands

No Comments

No Known Activations