INDEX

Explanations

themes related to intention and consequences in actions or objects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

ãĤ¥

-0.15

lename

-0.14

 blind

-0.14

NK

-0.14

 Offline

-0.14

_hdl

-0.14

">#

-0.14

cav

-0.14

 astr

-0.14

din

-0.13

POSITIVE LOGITS

 intentional

0.48

 deliberate

0.39

 intentionally

0.35

 intended

0.32

 purpose

0.32

 intention

0.31

 deliberately

0.28

purpose

0.28

 planned

0.28

 Purpose

0.27

Activations Density 0.225%

themes related to intention and consequences in actions or objects

No Comments

No Known Activations