INDEX

Explanations

phrases indicating acknowledgment or recognition of someone's qualities or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 reportedly

-0.22

 said

-0.19

said

-0.18

æį®

-0.18

 bekannt

-0.16

acco

-0.16

 ìķĮëł¤

-0.15

 ÑģÑĩÐ¸ÑĤÐ°ÐµÑĤÑģÑı

-0.15

 famously

-0.15

 æĺİ

-0.15

POSITIVE LOGITS

 indeed

0.48

inde

0.31

 Indeed

0.31

Indeed

0.30

 somehow

0.28

 headed

0.20

 ÙĪØ£ÙĨ

0.19

 actually

0.18

 worth

0.18

 heading

0.17

Activations Density 0.397%

phrases indicating acknowledgment or recognition of someone's qualities or actions

No Comments

No Known Activations