INDEX

Explanations

keywords and structures related to programming and user interface 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

arella

-0.17

 Hairst

-0.15

ROY

-0.15

 Pole

-0.14

ousel

-0.13

ement

-0.13

otten

-0.13

 Ø¨Ø¹

-0.13

_firestore

-0.13

 Underground

-0.13

POSITIVE LOGITS

Wunused

0.15

ongyang

0.14

enticate

0.14

tura

0.14

askell

0.14

 Crosby

0.14

alloca

0.14

 prof

0.14

.scalablytyped

0.14

ili

0.14

Activations Density 0.376%

keywords and structures related to programming and user interface components

No Comments

No Known Activations