INDEX

Explanations

instances of past and current service or leadership roles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

avers

-0.18

Services

-0.16

ervice

-0.16

å¾Ħ

-0.16

services

-0.15

 Services

-0.15

lou

-0.15

ixin

-0.14

æ´¥

-0.14

POSITIVE LOGITS

 notice

0.22

 alongside

0.20

illance

0.19

ably

0.18

 notices

0.16

notice

0.16

 noticing

0.16

/work

0.15

aname

0.15

 customers

0.15

Activations Density 0.028%

instances of past and current service or leadership roles

No Comments

No Known Activations