INDEX

Explanations

phrases formatted with specific textual styles such as bold or capital letters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

uzu

-0.15

un

-0.14

Ð¾Ð·Ð°

-0.14

ra

-0.14

ves

-0.14

Ð¾ÐºÐ¾Ð¼

-0.14

ª

-0.14

ÏģÎµ

-0.14

 struggles

-0.13

POSITIVE LOGITS

 bold

0.56

 capital

0.50

 Bold

0.46

 CAPITAL

0.46

bold

0.45

 capitals

0.45

Bold

0.43

capital

0.41

 italic

0.40

-bold

0.40

Activations Density 0.194%

phrases formatted with specific textual styles such as bold or capital letters

No Comments

No Known Activations