INDEX

Explanations

programming-related terms and operations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

Updates

-0.20

 asset

-0.18

 Formats

-0.18

Forms

-0.17

 Cases

-0.17

 ingredient

-0.17

Formats

-0.17

 item

-0.17

 Manuals

-0.17

Levels

-0.17

POSITIVE LOGITS

 objects

0.26

 cards

0.26

 tokens

0.25

 foods

0.25

 documents

0.25

 codes

0.25

 devices

0.24

 files

0.24

 services

0.24

 games

0.24

Activations Density 0.203%

programming-related terms and operations

No Comments

No Known Activations