INDEX

Explanations

elements related to predefined structures or setups within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ÑĢÐ°ÑīÐµÐ½Ð¸Ñı

-0.16

umpt

-0.16

usz

-0.14

abee

-0.14

reon

-0.14

 Pole

-0.14

OOM

-0.14

ato

-0.14

zem

-0.14

aliz

-0.14

POSITIVE LOGITS

:↵

0.38

:↵↵

0.34

ä»¥ä¸ĭ

0.34

å¦Ĥä¸ĭ

0.32

:č↵

0.32

 following

0.30

 ëĭ¤ìĿĮê³¼

0.30

ï¼ļ↵

0.28

():↵

0.28

 seguint

0.28

Activations Density 0.002%

elements related to predefined structures or setups within a context

No Comments

No Known Activations