INDEX

Explanations

key terms and phrases that indicate significant observations or findings

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

ington

-0.15

letcher

-0.15

Ì§

-0.14

oundation

-0.14

aos

-0.14

voj

-0.14

cons

-0.13

leared

-0.13

avier

-0.13

ÃŃda

-0.13

POSITIVE LOGITS

atz

0.16

0.15

emony

0.15

 Eins

0.14

sf

0.14

elin

0.14

 showc

0.14

ãĥ©ãĤ¹

0.13

erto

0.13

Activations Density 0.121%

key terms and phrases that indicate significant observations or findings

No Comments

No Known Activations