LLM EXPLORER 67,627 MODELS INDEXED

Pythia 160M Hq Emails by postbot

By postbot · 16 downloads

Pythia 160M Hq Emails is an open-source language model by postbot. Features: 160m LLM, VRAM: 0.7GB, Context: 2K, License: apache-2.0, LLM Explorer Score: 0.1, Arc: 23.1, HellaSwag: 30.1, MMLU: 26.6.

Base model:eleutherai/pythia-1... Base model:finetune:eleutherai... Dataset:postbot/multi-emails-h...   En   Endpoints compatible   Generated from trainer   Gpt neox   Model-index   Pytorch   Region:us   Safetensors

Pythia 160M Hq Emails Parameters and Internals

Model Type 
text generation
Additional Notes 
This model is a fine-tuned version achieving a loss of 2.2856, accuracy of 0.6113, and perplexity of 9.8313 on the evaluation set.
Training Details 
Data Sources:
postbot/multi-emails-hq
LLM NamePythia 160M Hq Emails
Repository πŸ€—https://huggingface.co/postbot/pythia-160m-hq-emails 
Base Model(s)  Pythia 160M Deduped   EleutherAI/pythia-160m-deduped
Model Size160m
Required VRAM0.7 GB
Updated2026-07-11
Maintainerpostbot
Model Typegpt_neox
Model Files  0.7 GB   0.7 GB   0.0 GB
Supported Languagesen
Model ArchitectureGPTNeoXForCausalLM
Licenseapache-2.0
Context Length2048
Model Max Length2048
Transformers Version4.27.0.dev0
Tokenizer ClassGPTNeoXTokenizer
Vocabulary Size50304
Torch Data Typefloat32

Best Alternatives to Pythia 160M Hq Emails

Best Alternatives
Context / RAM
Downloads
Likes
Pythia 160M C2s8K / 0.6 GB2517
Violet 160M Chat4K / 0.3 GB130
...50 Previous Mse Seed324 Stage12K / 0.2 GB9530
...50 Previous Mse Seed208 Stage12K / 0.2 GB3630
...0 Previous Mse Seed1024 Stage12K / 0.2 GB3590
Pythia 160M2K / 0.4 GB229417444
Pythia 160M Deduped2K / 0.4 GB1334373
Pythia 160M Xsum Roya2K / 0.6 GB60
Pythia 160m Sft2K / 0 GB80
Sheared Pythia 160M2K / 0.7 GB534
Note: green Score (e.g. "73.2") means that the model is better than postbot/pythia-160m-hq-emails.