INDEX

Explanations

geographical references and locations, particularly in relation to educational facilities or programs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

pheres

-0.16

acf

-0.15

aley

-0.15

aly

-0.15

Florida

-0.14

 Hurricane

-0.14

flip

-0.14

heit

-0.14

ç¨¿

-0.14

å±±å¸Ĥ

-0.14

POSITIVE LOGITS

 Kiss

0.26

 Brad

0.22

 Winter

0.21

 Plant

0.21

Imm

0.21

 Tavern

0.20

Pun

0.20

 Pemb

0.20

 Stuart

0.19

 Clear

0.19

Activations Density 0.049%

geographical references and locations, particularly in relation to educational facilities or programs

No Comments

No Known Activations