INDEX

Explanations

numbers and their formatting in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

itas

-0.16

 Hang

-0.14

unos

-0.14

âĢĮâĢĮ

-0.14

itis

-0.14

','','

-0.14

serter

-0.14

alian

-0.14

venge

-0.13

/mat

-0.13

POSITIVE LOGITS

usercontent

0.17

igel

0.15

oled

0.15

orent

0.15

Ð²ÑĸÑĤ

0.14

 Pelosi

0.14

ilan

0.14

usz

0.14

aina

0.13

bsolute

0.13

Activations Density 0.048%

numbers and their formatting in various contexts

No Comments

No Known Activations