INDEX

Explanations

instances of the word "re"

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

nt

-0.27

-0.26

ãģ¦ãģĦãĤĭ

-0.26

nd

-0.25

sWith

-0.25

ãģ¦

-0.25

POSITIVE LOGITS

iw

0.18

iros

0.18

xp

0.18

ngine

0.17

er

0.17

preneur

0.17

venue

0.17

vious

0.17

nger

0.16

an

0.16

Activations Density 0.018%

instances of the word "re"

No Comments

No Known Activations