INDEX

Explanations

references to the process of taking or transferring items, actions, or concepts from one context to another

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

_FOREACH

-0.16

leh

-0.15

orthy

-0.15

INGER

-0.14

.reserve

-0.14

edin

-0.14

-Mart

-0.14

aphael

-0.14

 ðŁĺī↵↵

-0.14

gMaps

-0.14

POSITIVE LOGITS

 thereafter

0.34

 then

0.27

 subsequent

0.27

ãģĿãģĵ

0.26

 subsequently

0.26

 further

0.24

 THEN

0.24

 next

0.23

 ensuite

0.22

 Ð´Ð°Ð»ÑĮ

0.21

Activations Density 0.159%

references to the process of taking or transferring items, actions, or concepts from one context to another

No Comments

No Known Activations