INDEX

Explanations

references to the word "Orange" 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 @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

edly

-0.21

ly

-0.20

artment

-0.16

woke

-0.16

lier

-0.16

atically

-0.15

loo

-0.15

erable

-0.15

Voy

-0.15

iland

-0.14

POSITIVE LOGITS

burg

0.26

utan

0.25

 Julius

0.22

 County

0.22

 peel

0.22

 juice

0.21

pek

0.20

 Peel

0.20

 blossom

0.20

ade

0.19

Activations Density 0.008%

references to the word "Orange" in various contexts

No Comments

No Known Activations