INDEX

Explanations

specific weight measurements and descriptions related to objects, particularly in terms of heaviness or lightness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ekli

-0.16

EPS

-0.15

argas

-0.15

eks

-0.15

rescia

-0.15

ekler

-0.15

Ãło

-0.14

uldu

-0.14

efore

-0.14

antar

-0.14

POSITIVE LOGITS

 weight

0.21

 weights

0.19

 Weight

0.18

weights

0.17

weight

0.17

zimmer

0.16

(weight

0.16

.scalablytyped

0.16

Weights

0.16

Weight

0.16

Activations Density 0.098%

specific weight measurements and descriptions related to objects, particularly in terms of heaviness or lightness

No Comments

No Known Activations