INDEX

Explanations

code-related functions and drawing instructions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 accordion

-0.17

column

-0.15

carousel

-0.15

GridLayout

-0.15

compress

-0.15

 Rows

-0.15

 insets

-0.15

columns

-0.14

ORIZONTAL

-0.14

Columns

-0.14

POSITIVE LOGITS

 line

0.29

 lines

0.28

 shape

0.27

 shapes

0.27

 Line

0.26

 drawn

0.25

 path

0.24

 Shape

0.24

rawn

0.24

 points

0.23

Activations Density 0.148%

code-related functions and drawing instructions

No Comments

No Known Activations