INDEX

Explanations

snippets of code or programming syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

iÅ¾

-0.17

ÃŃnh

-0.15

ÎµÏį

-0.15

@class

-0.15

aba

-0.14

ekk

-0.14

ike

-0.14

iets

-0.14

 Kingston

-0.14

FromClass

-0.14

POSITIVE LOGITS

 public

0.35

public

0.29

 void

0.28

 pÃºblico

0.23

	public

0.22

-public

0.22

 protected

0.22

 private

0.21

protected

0.19

 static

0.19

Activations Density 0.079%

snippets of code or programming syntax

No Comments

No Known Activations