INDEX

Explanations

phrases related to architecture and features of products

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

eters

-0.15

letes

-0.14

Ã©se

-0.14

akan

-0.14

hol

-0.14

ano

-0.14

kel

-0.13

anda

-0.13

datal

-0.13

Dat

-0.13

POSITIVE LOGITS

 Brilliant

0.20

 Elegant

0.18

 Desire

0.18

 Awesome

0.17

 Amazing

0.17

 Warm

0.17

Sty

0.16

At

0.16

 Contemporary

0.16

 Household

0.15

Activations Density 0.005%

phrases related to architecture and features of products

No Comments

No Known Activations