INDEX

Explanations

key concepts and best practices related to software architecture and design patterns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

æĶ¶å½ķ

-0.15

-prefix

-0.14

gre

-0.14

OptionsResolver

-0.13

ide

-0.13

_duplicates

-0.13

prd

-0.13

lsi

-0.13

ibri

-0.13

Gre

-0.13

POSITIVE LOGITS

 logic

0.20

 things

0.20

 stuff

0.19

 data

0.19

UserCode

0.18

 objects

0.18

 complex

0.18

stuff

0.17

aska

0.17

innen

0.17

Activations Density 0.308%

key concepts and best practices related to software architecture and design patterns

No Comments

No Known Activations