INDEX

Explanations

the word "well" 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 @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ynes

-0.24

ollo

-0.16

 Ð¿Ñĥ

-0.15

linger

-0.15

igs

-0.15

èŀº

-0.15

yne

-0.15

 nÃło

-0.14

elif

-0.14

 CALLBACK

-0.14

POSITIVE LOGITS

as

0.21

-known

0.20

 well

0.19

-being

0.17

spring

0.17

Äħd

0.17

/or

0.16

being

0.16

ãĤĪãģ³

0.15

 known

0.15

Activations Density 0.026%

the word "well" in various contexts

No Comments

No Known Activations