INDEX

Explanations

occurrences of the word "play" and its variations, indicating a focus on activities or engaging actions

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

oo

-0.15

idable

-0.15

uary

-0.15

endale

-0.14

 Shed

-0.14

idi

-0.14

mland

-0.14

igne

-0.14

 glBegin

-0.14

ri

-0.13

POSITIVE LOGITS

 host

0.23

 favorites

0.17

ibr

0.17

fully

0.17

ourd

0.17

 havoc

0.16

erv

0.15

oulos

0.15

vanized

0.15

buzz

0.15

Activations Density 0.012%

occurrences of the word "play" and its variations, indicating a focus on activities or engaging actions

No Comments

No Known Activations