INDEX

Explanations

conjunctions and particles indicating connections or relationships within text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 opposite

-0.15

epam

-0.14

 Lund

-0.14

efa

-0.14

map

-0.14

imedia

-0.14

 McKenzie

-0.14

efault

-0.14

map

-0.13

ingga

-0.13

POSITIVE LOGITS

teÅŁ

0.15

irs

0.15

 âĹĦ

0.14

ÑģÐºÐ»Ð°Ð´

0.14

IGN

0.13

phans

0.13

_sprites

0.13

apper

0.13

appoint

0.13

alette

0.13

Activations Density 0.256%

conjunctions and particles indicating connections or relationships within text

No Comments

No Known Activations