INDEX

Explanations

references to the term "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 @ 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

ly

-0.22

edly

-0.18

etch

-0.18

lya

-0.17

praak

-0.17

loo

-0.17

woke

-0.17

ubar

-0.16

LY

-0.16

atically

-0.15

POSITIVE LOGITS

utan

0.25

 juice

0.22

 peel

0.21

 Julius

0.21

burg

0.20

 Peel

0.20

 County

0.20

/red

0.18

 Juice

0.18

Activations Density 0.013%

references to the term "Orange" in various contexts

No Comments

No Known Activations

references to the term "Orange" in various contexts

No Comments

No Known Activations