INDEX

Explanations

phrases indicating potential assistance or support

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

anyak

-0.15

laz

-0.14

olec

-0.13

Slo

-0.13

rin

-0.13

IME

-0.13

.camel

-0.13

-âĢĲ

-0.13

oba

-0.13

ayan

-0.13

POSITIVE LOGITS

 better

0.24

better

0.22

 improvement

0.21

 mieux

0.21

 mejor

0.21

 melhor

0.20

 improved

0.20

 improve

0.19

 Ð»ÑĥÑĩÑĪÐµ

0.19

 Improvement

0.18

Activations Density 0.093%

phrases indicating potential assistance or support

No Comments

No Known Activations