INDEX

Explanations

references to individuals' professional roles and achievements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ëĤĺëĿ¼

-0.14

.crm

-0.14

.gf

-0.14

typeid

-0.13

\brief

-0.13

omba

-0.13

 participation

-0.13

ulus

-0.13

Æ°

-0.13

 involvement

-0.13

POSITIVE LOGITS

 speaker

0.31

 author

0.28

 Speakers

0.26

 speakers

0.26

 Speaker

0.26

Speaker

0.25

 speaking

0.25

peaker

0.25

 Author

0.24

 keynote

0.24

Activations Density 0.221%

references to individuals' professional roles and achievements

No Comments

No Known Activations