INDEX

Explanations

expressions related to personal experiences and reflections

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

ekim

-0.17

viar

-0.16

à¸Ļà¸²à¸¡

-0.16

ulumi

-0.15

kyt

-0.15

orris

-0.15

ll

-0.15

orr

-0.15

shall

-0.15

apur

-0.15

POSITIVE LOGITS

Lat

0.34

lat

0.29

Lat

0.28

 latent

0.26

 latency

0.26

 likely

0.26

LAT

0.25

 never

0.25

 probably

0.24

-lat

0.23

Activations Density 0.037%

expressions related to personal experiences and reflections

No Comments

No Known Activations