INDEX

Explanations

references to discussions and forums related to Windows and various technical support topics

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

ucken

-0.14

oplay

-0.14

Subscribe

-0.14

imeo

-0.14

uar

-0.13

Å¡ov

-0.13

 dilation

-0.13

anela

-0.13

_WRAP

-0.13

ndef

-0.13

POSITIVE LOGITS

 thread

0.33

 threads

0.31

 General

0.30

THREAD

0.28

 Thread

0.28

 general

0.28

thread

0.28

 Threads

0.26

threads

0.26

General

0.26

Activations Density 0.049%

references to discussions and forums related to Windows and various technical support topics

No Comments

No Known Activations