INDEX

Explanations

references to the word "Holland."

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

aina

-0.15

etable

-0.14

ÙĨØ§ÙĨ

-0.14

à¤łà¤¨

-0.14

.styleable

-0.14

Ð¿ÐµÑĩ

-0.14

ancybox

-0.14

ä¼ı

-0.13

etype

-0.13

ì½

-0.13

POSITIVE LOGITS

ìĦľëĬĶ

0.15

 Armstrong

0.15

reich

0.15

oders

0.15

odge

0.14

ishly

0.14

auge

0.14

OLL

0.13

det

0.13

Activations Density 0.003%

references to the word "Holland."

No Comments

No Known Activations