INDEX

Explanations

phrases questioning the efficacy or value of actions and their outcomes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

apes

-0.15

adr

-0.15

à¹ĥà¸Ī

-0.14

ALTH

-0.14

ÑģÐ¾Ð»

-0.14

ouncer

-0.14

incinn

-0.14

igure

-0.14

outh

-0.13

æ£

-0.13

POSITIVE LOGITS

affen

0.17

Įĵ

0.15

 Miche

0.15

 Benson

0.15

Bry

0.15

 Michaels

0.14

Ã³s

0.14

olo

0.14

Lup

0.14

 nominated

0.14

Activations Density 0.129%

phrases questioning the efficacy or value of actions and their outcomes

No Comments

No Known Activations