INDEX

Explanations

phrases or expressions that suggest a challenging or clichéd statement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

stin

-0.17

anism

-0.15

erland

-0.15

 geld

-0.15

Spinner

-0.15

 Judiciary

-0.14

benchmark

-0.14

precated

-0.14

/umd

-0.14

arding

-0.14

POSITIVE LOGITS

avo

0.17

-fashion

0.14

alendar

0.14

quet

0.14

 Armour

0.14

yme

0.13

ģ

0.13

rift

0.13

STEM

0.13

ä¼Ŀ

0.13

Activations Density 0.193%

phrases or expressions that suggest a challenging or clichéd statement

No Comments

No Known Activations