INDEX

Explanations

references to themes in 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 @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

aneous

-0.18

ree

-0.17

teen

-0.17

BE

-0.16

inet

-0.16

ardo

-0.15

anca

-0.15

ty

-0.15

hti

-0.15

 themed

-0.15

POSITIVE LOGITS

elves

0.24

atically

0.21

 park

0.20

 ÑģÐ°Ð¼ÑĭÐ¼

0.19

atical

0.18

æĿĲ

0.18

eting

0.18

asurement

0.18

icals

0.17

Activations Density 0.016%

references to themes in various contexts

No Comments

No Known Activations