INDEX

Explanations

references to specific individuals and their affiliations or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

ÙĴÙģ

-0.16

ffe

-0.15

adin

-0.14

Ø§ÙĪÛĮ

-0.14

wel

-0.14

Ãºi

-0.14

uristic

-0.14

pcion

-0.14

 seedu

-0.13

udded

-0.13

POSITIVE LOGITS

.pa

0.15

ills

0.15

arez

0.14

ulan

0.14

åİħ

0.13

/Page

0.13

utherland

0.13

zzo

0.13

ker

0.13

alc

0.12

Activations Density 0.344%

references to specific individuals and their affiliations or titles

No Comments

No Known Activations