INDEX

Explanations

references to services and their associated qualities or conditions

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

.netty

-0.16

ncia

-0.14

 Schwarz

-0.14

ixture

-0.14

leigh

-0.14

à¹Ĥà¸¢

-0.13

kas

-0.13

 entend

-0.13

åİĨ

-0.13

ewidth

-0.13

POSITIVE LOGITS

 services

1.04

services

0.85

 Services

0.85

Services

0.79

 SERVICES

0.76

-services

0.75

_services

0.72

ervices

0.70

 service

0.68

 ÑĥÑģÐ»ÑĥÐ³

0.68

Activations Density 0.252%

references to services and their associated qualities or conditions

No Comments

No Known Activations