INDEX

Explanations

roles and titles associated with positions of authority or governance

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

 alone

-0.16

edor

-0.14

.vn

-0.14

 which

-0.14

Î¬Î»

-0.14

Î²Îµ

-0.13

stÃ¤nd

-0.13

 Ú¯Ø³Øª

-0.13

oro

-0.13

 itself

-0.13

POSITIVE LOGITS

0.21

åħ¼

0.21

 Emer

0.20

–

0.19

0.18

cum

0.17

cum

0.17

/System

0.16

0.15

Activations Density 0.079%

roles and titles associated with positions of authority or governance

No Comments

No Known Activations