INDEX

Explanations

proper nouns, particularly names and institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ëĵĿ

-0.17

osten

-0.16

illon

-0.16

halt

-0.16

ainers

-0.15

eson

-0.15

assis

-0.15

Sus

-0.15

ainless

-0.15

ivan

-0.15

POSITIVE LOGITS

Fi

0.24

fi

0.22

rst

0.20

orent

0.20

Fi

0.18

-fi

0.18

allback

0.17

rist

0.17

sher

0.17

-Fi

0.17

Activations Density 0.013%

proper nouns, particularly names and institutions

No Comments

No Known Activations