INDEX

Explanations

references to television networks and programming

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

odcast

-0.17

isu

-0.15

stadt

-0.13

ÙĤÙĩ

-0.13

 Particle

-0.13

 Township

-0.13

iÄįky

-0.13

ptal

-0.13

.toObject

-0.13

vw

-0.13

POSITIVE LOGITS

 channel

0.40

 channels

0.38

 Channel

0.36

channel

0.35

channels

0.33

 Channels

0.33

-channel

0.33

 CHANNEL

0.31

 cable

0.31

Channel

0.29

Activations Density 0.206%

references to television networks and programming

No Comments

No Known Activations