INDEX

Explanations

words related to revealing or introducing new products or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ovice

-0.15

anes

-0.15

ìłĪ

-0.15

endl

-0.15

ownt

-0.15

xima

-0.15

linear

-0.14

alamat

-0.14

 Rowe

-0.14

irie

-0.14

POSITIVE LOGITS

ils

0.23

veal

0.17

al

0.17

0.16

illed

0.16

ek

0.16

aul

0.16

er

0.16

ials

0.15

illing

0.15

Activations Density 0.003%

words related to revealing or introducing new products or concepts

No Comments

No Known Activations