INDEX

Explanations

references to volume and measurement metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

estro

-0.18

 Dion

-0.17

emente

-0.15

æĮ¯ãĤĬ

-0.15

izable

-0.15

hood

-0.15

Ð¾ÑĢÐ°

-0.15

ernote

-0.14

ams

-0.14

igkeit

-0.14

POSITIVE LOGITS

untary

0.27

unteer

0.25

atility

0.24

atile

0.23

taire

0.23

unteers

0.22

swagen

0.21

uble

0.20

leys

0.20

cano

0.20

Activations Density 0.016%

references to volume and measurement metrics

No Comments

No Known Activations