INDEX

Explanations

references to historical architecture and structures

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

oras

-0.17

-0.14

bjerg

-0.14

uru

-0.14

Sector

-0.14

ordon

-0.14

aniel

-0.14

 Sector

-0.14

á»ĳ

-0.13

POSITIVE LOGITS

 Carol

0.25

 Frank

0.25

 Franken

0.23

Frank

0.22

 Lomb

0.21

Pep

0.18

 pÅĻem

0.18

 Conrad

0.17

 PÅĻ

0.17

 Worm

0.17

Activations Density 0.032%

references to historical architecture and structures

No Comments

No Known Activations