INDEX

Explanations

repeated phrases or ideas that indicate familial relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 3-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L3R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.3.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

doz

-0.14

 better

-0.14

rips

-0.14

 localtime

-0.14

bor

-0.14

Luz

-0.14

vin

-0.14

-heavy

-0.14

 Helm

-0.13

ulia

-0.13

POSITIVE LOGITS

ataka

0.17

ãģ¬

0.15

lander

0.15

birth

0.14

igs

0.14

htag

0.14

ÏģÎŃ

0.14

mai

0.14

Photos

0.13

ayar

0.13

Activations Density 0.010%

repeated phrases or ideas that indicate familial relationships

No Comments

No Known Activations