INDEX

Explanations

words indicating the introduction of information or a list

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

éħ¸

-0.16

istÃ©

-0.16

affer

-0.15

ATAR

-0.15

.sharedInstance

-0.14

_MSK

-0.14

éłŃ

-0.14

failure

-0.14

æĸĹ

-0.14

Jun

-0.14

POSITIVE LOGITS

egra

0.16

orde

0.16

arry

0.16

Tal

0.15

anik

0.15

istle

0.15

 Gran

0.14

 Ø¨ÙĪØ§Ø¨Ø©

0.14

 Equipment

0.14

Kur

0.14

Activations Density 0.000%

words indicating the introduction of information or a list

No Comments

No Known Activations