INDEX

Explanations

phrases related to obligations and responsibilities

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

isque

-0.17

ooled

-0.15

Sensitive

-0.15

_sent

-0.15

 Sent

-0.15

 embody

-0.14

 sent

-0.14

Sent

-0.14

 drafted

-0.14

 SENT

-0.14

POSITIVE LOGITS

 respected

0.26

met

0.26

 cater

0.23

 honored

0.22

 fulfilled

0.20

met

0.20

 accounted

0.20

 addressed

0.19

 attended

0.19

 known

0.19

Activations Density 0.162%

phrases related to obligations and responsibilities

No Comments

No Known Activations