INDEX

Explanations

temporal markers or indicators of time in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

affer

-0.17

bach

-0.16

zan

-0.16

zag

-0.15

ug

-0.15

acker

-0.15

 hall

-0.15

 upstream

-0.15

è¤

-0.14

 padd

-0.14

POSITIVE LOGITS

wards

0.32

words

0.31

ward

0.31

 wards

0.29

æŃ¤

0.24

 ÄĳÃ³

0.22

WARD

0.22

 Ø°ÙĦÙĥ

0.21

WARDS

0.20

word

0.20

Activations Density 0.152%

temporal markers or indicators of time in the text

No Comments

No Known Activations