INDEX

Explanations

expressions of excitement and engagement in social interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

tero

-0.17

pedia

-0.15

ÑĨÐ¾Ð²

-0.15

eyin

-0.15

aub

-0.15

angu

-0.14

KEN

-0.14

terior

-0.14

YOU

-0.14

 táº§ng

-0.14

POSITIVE LOGITS

 would

0.15

Ð¸ÑĤÐ¾Ð¼

0.14

entries

0.14

odom

0.14

noch

0.13

entry

0.13

 Tough

0.13

Activations Density 0.002%

expressions of excitement and engagement in social interactions

No Comments

No Known Activations