INDEX

Explanations

terms related to dominance and control in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

uman

-0.19

tract

-0.17

Åº

-0.17

ses

-0.17

ocation

-0.17

sl

-0.16

 dozen

-0.16

burn

-0.15

enia

-0.15

riott

-0.15

POSITIVE LOGITS

 ÑģÐ¾Ð±Ð¾Ð¹

0.19

estic

0.17

ERRU

0.17

/dom

0.16

 proceedings

0.16

 headlines

0.16

 ÑģÐ¾Ð±Ð¾Ñİ

0.15

antly

0.15

ascar

0.15

/control

0.15

Activations Density 0.035%

terms related to dominance and control in various contexts

No Comments

No Known Activations