INDEX

Explanations

references to Pokémon and related terminology

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

gerald

-0.17

.scalablytyped

-0.16

Gam

-0.16

anship

-0.15

enguin

-0.14

ez

-0.14

omore

-0.14

uin

-0.14

POSITIVE LOGITS

Ã©mon

0.24

itto

0.19

emons

0.18

Ã©d

0.18

EMON

0.17

Ã©g

0.17

Pok

0.17

orny

0.17

mon

0.16

olen

0.15

Activations Density 0.010%

references to Pokémon and related terminology

No Comments

No Known Activations