INDEX

Explanations

terms related to hockey

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 Leban

-0.16

abr

-0.15

/qt

-0.15

 Benton

-0.14

atatype

-0.14

 Bris

-0.14

AAF

-0.14

Åº

-0.14

 Auburn

-0.14

 kÃª

-0.14

POSITIVE LOGITS

 Hockey

0.38

 hockey

0.35

FI

0.32

FI

0.26

 Argentina

0.23

Argentina

0.21

 Penalty

0.21

 stick

0.20

 circle

0.20

 Frozen

0.20

Activations Density 0.015%

terms related to hockey

No Comments

No Known Activations