INDEX

Explanations

instances of text labeled as "written" followed by an author or attribution

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

éº»

-0.17

 Ð½Ð°ÑĢ

-0.14

eed

-0.14

zu

-0.14

dale

-0.14

_mono

-0.14

iev

-0.14

 universal

-0.14

iei

-0.14

POSITIVE LOGITS

lest

0.20

Ã¤s

0.17

Ã¼f

0.15

 è¢

0.15

à¸©

0.15

IRROR

0.14

çĽ

0.14

 Ð¾Ð±ÑĬ

0.14

iasi

0.14

alsa

0.14

Activations Density 0.004%

instances of text labeled as "written" followed by an author or attribution

No Comments

No Known Activations