INDEX

Explanations

elements related to unique historical events or unusual occurrences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

-0.16

-0.15

Disclosure

-0.14

ixel

-0.14

Vul

-0.14

enga

-0.14

andon

-0.14

 Downing

-0.14

POSITIVE LOGITS

 Seeder

0.17

conto

0.15

.AttributeSet

0.15

etAddress

0.15

 geschichten

0.15

à¸´à¸į

0.15

iosper

0.14

åĳ³

0.14

 ëķ

0.14

ewolf

0.13

Activations Density 0.259%

elements related to unique historical events or unusual occurrences

No Comments

No Known Activations