INDEX

Explanations

references to parameters, especially those related to email and user actions in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

-p

-0.30

Âłp

-0.30

.p

-0.27

pd

-0.27

/pp

-0.27

[p

-0.26

.getP

-0.26

 getP

-0.26

pa

-0.25

,p

-0.25

POSITIVE LOGITS

Pipe

0.23

 Pipe

0.20

Pause

0.19

 Prompt

0.19

 Purpose

0.18

Percent

0.18

 Pill

0.18

Purpose

0.18

 Percent

0.17

_pipe

0.17

Activations Density 0.072%

references to parameters, especially those related to email and user actions in a programming context

No Comments

No Known Activations