INDEX

Explanations

words and phrases related to Android programming and its components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

inces

-0.16

izr

-0.15

agog

-0.15

 Sect

-0.15

 Kills

-0.15

_MAG

-0.15

 radial

-0.14

ÑĤÐ¸

-0.14

æĵ

-0.14

TIM

-0.14

POSITIVE LOGITS

.opensource

0.18

cko

0.18

 addCriterion

0.18

bÃ©

0.15

taj

0.15

aus

0.15

bette

0.15

.djang

0.15

och

0.14

Activations Density 0.001%

words and phrases related to Android programming and its components

No Comments

No Known Activations