INDEX

Explanations

phrases related to division or categorization

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

umd

-0.16

statt

-0.15

ÑģÐ¾Ð²

-0.15

Ã³sito

-0.14

 SUBSTITUTE

-0.14

_ASYNC

-0.14

ICY

-0.14

 CONTRIBUTORS

-0.14

gren

-0.14

ÑĢ

-0.14

POSITIVE LOGITS

two

0.24

 three

0.18

 several

0.17

å¦Ĥä¸ĭ

0.16

two

0.15

 thirds

0.15

 macros

0.15

èĮĥ

0.15

 separate

0.15

afia

0.15

Activations Density 0.028%

phrases related to division or categorization

No Comments

No Known Activations