INDEX

Explanations

expressions of opinions regarding the quality of experiences or things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

orny

-0.15

inen

-0.15

ãĥ¼ãĤ¿ãĥ¼

-0.14

 occasions

-0.14

emen

-0.14

zza

-0.14

aktu

-0.13

 ourselves

-0.13

olars

-0.13

obra

-0.13

POSITIVE LOGITS

 thing

0.33

 ever

0.28

 EVER

0.28

 anyone

0.23

 Thing

0.23

ever

0.23

 Ever

0.23

-ever

0.22

 coisa

0.21

thing

0.21

Activations Density 0.077%

expressions of opinions regarding the quality of experiences or things

No Comments

No Known Activations