INDEX

Explanations

references to downloadable content and digital resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

onest

-0.17

okino

-0.17

adia

-0.17

ardo

-0.14

Nas

-0.14

erc

-0.14

amera

-0.14

 trÆ°á»Łng

-0.14

etu

-0.14

inge

-0.14

POSITIVE LOGITS

 wall

0.19

-save

0.18

 save

0.18

 saving

0.18

-wall

0.18

Wall

0.17

 Save

0.17

 SAVE

0.17

save

0.16

fty

0.16

Activations Density 0.009%

references to downloadable content and digital resources

No Comments

No Known Activations