INDEX

Explanations

terms associated with episodic content and series in entertainment media

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

_representation

-0.14

ÙĩØ±Ùĩ

-0.14

 Venue

-0.14

BarButton

-0.13

dorf

-0.13

 germ

-0.13

Ã©v

-0.13

 exact

-0.13

ä»

-0.13

POSITIVE LOGITS

 drop

0.18

dro

0.17

 entitled

0.16

Dro

0.16

 chronic

0.15

drop

0.15

 titled

0.15

deb

0.15

eatures

0.14

_logging

0.14

Activations Density 0.063%

terms associated with episodic content and series in entertainment media

No Comments

No Known Activations