INDEX

Explanations

terms related to opinions and perspectives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

erk

-0.19

Ð²Ð¸

-0.17

lsi

-0.17

ampion

-0.15

aways

-0.15

foot

-0.14

arin

-0.14

aturdays

-0.14

ilder

-0.14

elp

-0.14

POSITIVE LOGITS

naire

0.18

ably

0.18

naires

0.17

sher

0.17

IRMWARE

0.16

/tutorial

0.16

aires

0.15

ãĤ·ãĥ¼

0.14

airs

0.14

Activations Density 0.041%

terms related to opinions and perspectives

No Comments

No Known Activations