INDEX

Explanations

instances of the reflexive verb "se" indicating actions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

se

-0.22

-0.17

ser

-0.17

se

-0.17

-0.15

Ð¾ÑĤÑĮ

-0.15

 forn

-0.14

chein

-0.14

son

-0.14

POSITIVE LOGITS

 udÃ¡l

0.24

cciÃ³n

0.21

ccion

0.20

cciones

0.20

bastian

0.19

jong

0.19

vero

0.18

ules

0.18

leccion

0.17

bast

0.16

Activations Density 0.013%

instances of the reflexive verb "se" indicating actions or states

No Comments

No Known Activations