INDEX

Explanations

references to usefulness or practical applications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

emet

-0.17

ening

-0.17

edBy

-0.16

ees

-0.16

eler

-0.16

edImage

-0.16

es

-0.15

eds

-0.15

anna

-0.15

ee

-0.15

POSITIVE LOGITS

itarian

0.44

ties

0.29

itar

0.27

izing

0.25

izzo

0.25

ty

0.25

ising

0.24

ization

0.23

izations

0.22

ized

0.22

Activations Density 0.004%

references to usefulness or practical applications

No Comments

No Known Activations