INDEX

Explanations

phrases indicating dissatisfaction or complaints

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

 Heck

-0.17

thal

-0.16

çī¹åĪ«

-0.15

imore

-0.15

ONO

-0.14

Ð±Ð¾

-0.14

ÐĶÐĲ

-0.14

 íĬ¹íŀĪ

-0.14

heck

-0.14

 heck

-0.14

POSITIVE LOGITS

 sounds

0.31

 Sounds

0.30

Sounds

0.28

 Translation

0.25

sounds

0.24

Translation

0.23

 translation

0.23

 sound

0.22

 nice

0.21

ounds

0.20

Activations Density 0.318%

phrases indicating dissatisfaction or complaints

No Comments

No Known Activations