INDEX

Explanations

references to the Pittsburgh Steelers football team and its players

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

NZ

-0.15

ecta

-0.15

adem

-0.15

 Thames

-0.15

ãĤ¦ãĥĪ

-0.15

allon

-0.14

ROP

-0.14

chedulers

-0.14

 Wand

-0.14

adh

-0.14

POSITIVE LOGITS

 Pittsburgh

0.68

 Steelers

0.55

 Pitt

0.49

 Alle

0.47

PIT

0.45

Alle

0.43

 Penguins

0.41

 Steel

0.39

 Carnegie

0.39

Steel

0.38

Activations Density 0.059%

references to the Pittsburgh Steelers football team and its players

No Comments

No Known Activations