INDEX

Explanations

elements related to graphics and visual data in the text

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

ãĥ©ãĤ¹

-0.15

Ã©l

-0.15

åł

-0.14

indow

-0.14

egrator

-0.14

iteli

-0.14

Î¼Î±ÏĦÎ¿

-0.14

stroy

-0.14

Liz

-0.14

Î³ÎµÎ½

-0.14

POSITIVE LOGITS

 scale

0.26

 trim

0.24

trim

0.21

 Scale

0.21

scale

0.21

 scales

0.20

 Trim

0.20

_scale

0.20

 clip

0.19

(scale

0.19

Activations Density 0.010%

elements related to graphics and visual data in the text

No Comments

No Known Activations