INDEX

Explanations

references to official roles or positions in government contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

Ã³st

-0.14

ente

-0.14

ullen

-0.14

argent

-0.14

CLU

-0.14

Ð½Ð°ÑģÐ»ÑĸÐ´Ð¾Ðº

-0.13

ãĥ³ãĥĹ

-0.13

wet

-0.13

Ø§Ø¸

-0.13

entic

-0.13

POSITIVE LOGITS

 underline

0.20

underscore

0.20

 stresses

0.18

 underscore

0.18

 stressed

0.17

 exchanged

0.17

 Stress

0.17

 exchanges

0.17

.exchange

0.17

 stress

0.16

Activations Density 0.025%

references to official roles or positions in government contexts

No Comments

No Known Activations