INDEX

Explanations

references to unique or specific categories or types within a context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

annel

-0.17

pau

-0.16

erva

-0.15

 KÃ¼r

-0.15

 ÄĳÃ¢y

-0.15

ipp

-0.14

porto

-0.14

ÃŃm

-0.14

-âĢĲ

-0.14

LETE

-0.13

POSITIVE LOGITS

its

0.39

its

0.30

Its

0.29

åħ¶

0.28

Its

0.27

his

0.26

 seus

0.25

sua

0.25

 their

0.24

seu

0.24

Activations Density 0.140%

references to unique or specific categories or types within a context

No Comments

No Known Activations