INDEX

Explanations

references to cinema and film

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

resse

-0.17

oenix

-0.15

issement

-0.15

-0.14

osit

-0.14

itan

-0.14

aliyet

-0.14

ìĹŃ

-0.14

adiator

-0.14

POSITIVE LOGITS

ematic

0.32

emat

0.27

emas

0.24

ERGY

0.23

ematics

0.22

EMA

0.22

ergy

0.21

Cin

0.21

cin

0.19

eworld

0.19

Activations Density 0.007%

references to cinema and film

No Comments

No Known Activations