INDEX

Explanations

phrases that reflect desires or wants related to actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

opsis

-0.16

ç§Ģ

-0.15

 tall

-0.14

@show

-0.14

ainment

-0.13

éĢ¸

-0.13

ÑĢÐ°Ð¿

-0.13

 Ukra

-0.13

Ý

-0.13

POSITIVE LOGITS

 Fell

0.15

illy

0.15

wnd

0.15

ãĥĥãĤ°

0.15

cel

0.15

ì§Ħ

0.15

 componentDid

0.14

stuff

0.14

ady

0.14

 Courtney

0.14

Activations Density 0.193%

phrases that reflect desires or wants related to actions

No Comments

No Known Activations