INDEX

Explanations

terminology related to measurements or averages

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

ÏĥÎ®

-0.18

.openg

-0.16

 nackte

-0.15

eil

-0.14

arov

-0.14

ÐµÐºÑĤÐ¸Ð²

-0.14

 slogans

-0.14

Ã¥r

-0.14

iÅ¡

-0.14

POSITIVE LOGITS

iant

0.17

ips

0.15

bery

0.15

igator

0.15

odal

0.15

Dra

0.15

||||

0.15

ora

0.15

abis

0.15

ber

0.14

Activations Density 0.049%

terminology related to measurements or averages

No Comments

No Known Activations