INDEX

Explanations

references to architectural styles or historical monuments

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

inka

-0.17

utsch

-0.17

RowAt

-0.14

.IContainer

-0.14

 Ð´Ð²Ð¸Ð¶

-0.14

urtle

-0.14

 Jungle

-0.14

 sÃ¼r

-0.13

 Ridley

-0.13

 Ð²Ð¸Ð±

-0.13

POSITIVE LOGITS

 reform

0.23

 tapi

0.20

 Reform

0.19

 rehabilit

0.19

 remodel

0.19

 rectangular

0.19

 decor

0.18

 coron

0.18

 ornament

0.18

 constr

0.17

Activations Density 0.015%

references to architectural styles or historical monuments

No Comments

No Known Activations