Downloads · 30 days
0
anlp-a1-trial1
AN5517/anlp-a1-trial1
anlp-a1-trial1 is a machine learning model from AN5517. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
This repository contains trained model checkpoints and evaluation outputs for ANLP Assignment 1 across five controlled architectural configurations:
Access
Public
Updated Sep 3, 2026
Repo size
492 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pt196 MB · 100%
At a glance
- License
- mit
- Access
- Public
- Created
- Sep 3, 2026
- Updated
- Sep 3, 2026
- SHA
- f1ea7354
Model card
From the Hugging Face model README
ANLP Assignment 1: Transformers from Scratch & Ablation Study
This repository contains trained model checkpoints and evaluation outputs for ANLP Assignment 1 across five controlled architectural configurations:
- C1 (Base): Sinusoidal Absolute PE + Multi-Head Attention (MHA) + Pre-LayerNorm + Scratch BPE
- C2 (RoPE): Rotary Positional Embedding (RoPE) + MHA + Pre-LayerNorm + Scratch BPE
- C3 (GQA): Sinusoidal Absolute PE + Grouped-Query Attention (GQA) + Pre-LayerNorm + Scratch BPE
- C4 (RMSNorm): Sinusoidal Absolute PE + MHA + RMSNorm + Scratch BPE
- C5 (BLT): Sinusoidal Absolute PE + MHA + Pre-LayerNorm + Token-Free Byte Latent Transformer
Experimental Results Summary
----------------------------------------------------------------------------------------------------------------
ANLP ASSIGNMENT 1 — ABLATION STUDY RESULTS
----------------------------------------------------------------------------------------------------------------
Config | Status | Val Loss | Seq Acc | Bit Acc | Levenshtein | BLEU | VRAM (GB) | Time (s)
----------------------------------------------------------------------------------------------------------------
C1 | SUCCESS | 5.7553 | 0.0% | 0.0% | 561.8 | 0.00 | 0.46 | 3.1
C5 | SUCCESS | 4.3724 | 0.0% | 20.9% | 460.6 | N/A | 0.24 | 2.8
----------------------------------------------------------------------------------------------------------------
- License
- mit
- Languages
- en
- Created
- Sep 3, 2026
- Updated
- Sep 3, 2026