INDEX

Explanations

titles and positions of authority within organizational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

INA

-0.18

esk

-0.16

islav

-0.15

utra

-0.14

jon

-0.14

 interpol

-0.14

anse

-0.14

hire

-0.13

.netbeans

-0.13

defgroup

-0.13

POSITIVE LOGITS

 External

0.26

 external

0.25

External

0.24

 operations

0.23

 strategy

0.22

external

0.21

_external

0.21

 communications

0.20

 corporate

0.19

 Operations

0.19

Activations Density 0.090%

titles and positions of authority within organizational contexts

No Comments

No Known Activations