INDEX

Explanations

phrases that indicate focus or attention towards a subject or topic

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

PT

-0.15

hest

-0.15

ause

-0.14

zim

-0.14

ÑĢÐ¾ÑĤÐ¸Ð²

-0.14

nal

-0.14

@nate

-0.14

AVE

-0.14

ãĤīãģı

-0.14

colo

-0.13

POSITIVE LOGITS

Tow

0.16

 toward

0.15

à¹Ģà¸Ļ

0.15

naÄį

0.15

 areas

0.14

how

0.14

uzzi

0.14

lix

0.13

olan

0.13

Activations Density 0.056%

phrases that indicate focus or attention towards a subject or topic

No Comments

No Known Activations