INDEX

Explanations

instances of the word "stop" and related forms, indicating pauses or interruptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

Ð´Ð°Ð²

-0.17

ichten

-0.16

urally

-0.15

ensing

-0.15

ureau

-0.15

ervals

-0.15

ÑĢÐ¸Ð²

-0.15

tos

-0.15

cling

-0.14

stride

-0.14

POSITIVE LOGITS

page

0.26

lights

0.25

gap

0.24

pered

0.22

per

0.22

-motion

0.22

/start

0.22

over

0.20

light

0.19

PING

0.19

Activations Density 0.024%

instances of the word "stop" and related forms, indicating pauses or interruptions

No Comments

No Known Activations