INDEX

Explanations

programming-related concepts and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

 lobby

-0.16

arget

-0.15

ippy

-0.15

plate

-0.14

zyst

-0.13

oi

-0.13

lein

-0.13

wake

-0.13

illy

-0.13

imits

-0.13

POSITIVE LOGITS

 procedure

0.28

procedure

0.27

 writeln

0.24

{$

0.22

riteln

0.22

{$

0.22

 procedures

0.21

 Procedure

0.21

 begin

0.21

Procedure

0.20

Activations Density 0.004%

programming-related concepts and structures

No Comments

No Known Activations