INDEX

Explanations

references to podcasts and their content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

 Filme

-0.16

 PureComponent

-0.16

TVs

-0.15

DVD

-0.14

ihan

-0.14

à¸¸à¹ī

-0.14

dvd

-0.14

 filme

-0.14

DVD

-0.14

uthor

-0.13

POSITIVE LOGITS

 podcast

0.59

pod

0.57

pod

0.57

Pod

0.56

Pod

0.55

_pod

0.53

 Podcast

0.51

 podcasts

0.51

 pods

0.45

.Pod

0.43

Activations Density 0.076%

references to podcasts and their content

No Comments

No Known Activations