INDEX

Explanations

references to academic publications and journals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ikon

-0.18

illian

-0.15

olland

-0.15

amo

-0.14

abol

-0.14

 Material

-0.14

prox

-0.13

 prox

-0.13

ucks

-0.13

 Matth

-0.13

POSITIVE LOGITS

Ap

0.22

ar

0.21

Ap

0.21

 Nature

0.19

PAS

0.19

 astro

0.19

 nature

0.18

 Astronomy

0.17

skyt

0.17

Nature

0.17

Activations Density 0.005%

references to academic publications and journals

No Comments

No Known Activations