INDEX

Explanations

conjunctions and phrases that signal connections or relationships among items or ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

-0.14

ouns

-0.14

oz

-0.13

se

-0.13

Ã½

-0.13

bourg

-0.13

 ìŀĪê³ł

-0.13

 ÄĮer

-0.13

-0.12

POSITIVE LOGITS

arra

0.15

RouterModule

0.15

Ð¾Ð»Ð¾Ð´

0.14

 Lastly

0.14

abin

0.14

 pÅĻÃŃpadnÄĽ

0.14

_PULL

0.14

_pb

0.14

 finally

0.13

æĪ

0.13

Activations Density 0.169%

conjunctions and phrases that signal connections or relationships among items or ideas

No Comments

No Known Activations