INDEX

Explanations

instances of distinction or separation between entities or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

onium

-0.15

ibraltar

-0.15

onomies

-0.15

 Fallon

-0.14

ÏīÏĥÎ·

-0.14

StatusLabel

-0.14

 Mend

-0.14

lai

-0.14

uhe

-0.14

ÐºÐ°

-0.13

POSITIVE LOGITS

 separate

0.24

 Separate

0.22

 seperate

0.21

 separately

0.20

çĭ¬ç«ĭ

0.18

 distinct

0.18

 independent

0.18

 independently

0.17

 separ

0.16

inton

0.15

Activations Density 0.142%

instances of distinction or separation between entities or concepts

No Comments

No Known Activations