INDEX

Explanations

numerical information and addresses in a specific format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

.scalablytyped

-0.23

.bc

-0.18

Ø¯ÙĪØ¯

-0.15

 fragmentation

-0.15

cliffe

-0.15

Raq

-0.14

 Colombian

-0.14

ĩ´

-0.14

LOPT

-0.14

ÑĩÐ¸Ð»

-0.14

POSITIVE LOGITS

 Iowa

0.53

 Moines

0.33

 Sioux

0.32

IA

0.25

owa

0.24

 Waterloo

0.24

 Omaha

0.23

Des

0.22

 Nebraska

0.22

 Ames

0.20

Activations Density 0.035%

numerical information and addresses in a specific format

No Comments

No Known Activations