INDEX

Explanations

references to studies, including academic citations and the year of publication

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ething

-0.19

ston

-0.15

ç¸

-0.14

owler

-0.14

 ÙħØ¹

-0.14

itecture

-0.14

 minority

-0.14

Ø¨Ø¯

-0.13

POSITIVE LOGITS

dish

0.17

Ð¾ÑĢÐ¾Ð´

0.15

">//

0.14

aeda

0.14

ovich

0.14

ãĥ¥ãĥ¼

0.14

zure

0.13

ffi

0.13

dsn

0.13

quina

0.13

Activations Density 0.019%

references to studies, including academic citations and the year of publication

No Comments

No Known Activations