INDEX

Explanations

terms related to comfort and convenience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

hung

-0.18

Åº

-0.18

 comfortable

-0.17

owered

-0.16

eger

-0.16

icap

-0.16

statt

-0.16

ovah

-0.16

traction

-0.16

ched

-0.16

POSITIVE LOGITS

ably

0.35

able

0.33

ingly

0.29

ability

0.28

 zone

0.28

 zones

0.27

ab

0.25

abl

0.24

ABLE

0.23

abel

0.23

Activations Density 0.024%

terms related to comfort and convenience

No Comments

No Known Activations