INDEX

Explanations

terms related to high-profile individuals and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ixo

-0.15

azar

-0.15

aaS

-0.15

ernals

-0.15

.MSG

-0.15

 cosa

-0.15

.spy

-0.14

[assembly

-0.14

oundingBox

-0.14

Æ°á»Łng

-0.14

POSITIVE LOGITS

aight

0.17

scar

0.16

atus

0.15

gab

0.14

assen

0.14

VED

0.14

tec

0.14

ed

0.13

 Vince

0.13

gang

0.13

Activations Density 0.004%

terms related to high-profile individuals and events

No Comments

No Known Activations