INDEX

Explanations

phrases that emphasize comprehensiveness and thoroughness across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

jav

-0.15

Ã¤ch

-0.15

íĨł

-0.15

stro

-0.14

aminer

-0.14

inus

-0.14

iac

-0.14

ÐĵÐŀ

-0.14

-0.13

 mass

-0.13

POSITIVE LOGITS

 everything

0.23

everything

0.19

ä¸ĢåĪĩ

0.18

 tudo

0.17

 except

0.16

Except

0.16

Everything

0.16

except

0.16

ertz

0.16

 Everything

0.15

Activations Density 0.182%

phrases that emphasize comprehensiveness and thoroughness across various contexts

No Comments

No Known Activations