INDEX

Explanations

negative contractions or phrases that indicate disagreement or lack of ability

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

auses

-0.16

Ð°Ð»Ñĥ

-0.15

-valu

-0.15

htub

-0.15

anes

-0.15

achuset

-0.14

itory

-0.14

 authDomain

-0.14

ntax

-0.13

 ÑģÑĤÐ°ÑĢÐ°

-0.13

POSITIVE LOGITS

 ever

0.30

 EVER

0.26

ever

0.24

-ever

0.24

 anymore

0.24

 necessarily

0.23

 mind

0.22

 unless

0.22

 Ever

0.20

 anytime

0.20

Activations Density 0.102%

negative contractions or phrases that indicate disagreement or lack of ability

No Comments

No Known Activations