INDEX

Explanations

mentions of podcasts

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

esco

-0.16

oire

-0.15

æĿ

-0.15

orado

-0.15

MBER

-0.14

ptal

-0.14

AMA

-0.14

ringe

-0.14

á»¹

-0.14

odesk

-0.14

POSITIVE LOGITS

casts

0.24

iatric

0.21

cast

0.21

ium

0.19

 cast

0.19

iat

0.18

Pod

0.18

.Pod

0.18

 Save

0.17

 casts

0.17

Activations Density 0.013%

mentions of podcasts

No Comments

No Known Activations