INDEX

Explanations

references to variable assignments and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

á»Ļ

-0.17

æĹı

-0.17

ä¸ĸ

-0.16

emmel

-0.16

nees

-0.15

duk

-0.15

çķ

-0.15

FileSystem

-0.14

linger

-0.14

allen

-0.14

POSITIVE LOGITS

0.17

;amp

0.15

tif

0.15

0.14

.toString

0.14

.ToString

0.14

 \↵

0.14

 +↵

0.14

0.13

amment

0.13

Activations Density 0.056%

references to variable assignments and operations

No Comments

No Known Activations