|
---
|
|
license: apache-2.0
|
|
pipeline_tag: text-generation
|
|
library_name: transformers
|
|
language: [
|
|
'en', 'am', 'ar', 'as', 'az', 'be', 'bg', 'bn', 'br', 'bs', 'ca', 'cs', 'cy', 'da', 'de', 'el',
|
|
'eo', 'es', 'et', 'eu', 'fa', 'ff', 'fi', 'fr', 'fy', 'ga', 'gd', 'gl', 'gn', 'gu', 'ha', 'he',
|
|
'hi', 'hr', 'ht', 'hu', 'hy', 'id', 'ig', 'is', 'it', 'ja', 'jv', 'ka', 'kk', 'km', 'kn', 'ko',
|
|
'ku', 'ky', 'la', 'lg', 'li', 'ln', 'lo', 'lt', 'lv', 'mg', 'mk', 'ml', 'mn', 'mr', 'ms', 'my',
|
|
'ne', 'nl', 'no', 'ns', 'om', 'or', 'pa', 'pl', 'ps', 'pt', 'qu', 'rm', 'ro', 'ru', 'sa', 'si',
|
|
'sc', 'sd', 'sk', 'sl', 'so', 'sq', 'sr', 'ss', 'su', 'sv', 'sw', 'ta', 'te', 'th', 'tl', 'tn',
|
|
'tr', 'ug', 'uk', 'ur', 'uz', 'vi', 'wo', 'xh', 'yi', 'yo', 'zu',
|
|
]
|
|
datasets: [
|
|
'yahma/alpaca-cleaned',
|
|
'gbharti/wealth-alpaca_lora',
|
|
'saillab/taco-datasets',
|
|
'xu-song/cc100-samples',
|
|
'ontocord/fineweb-permissive-multilingual-2m',
|
|
'MuskumPillerum/General-Knowledge',
|
|
'yirenc/general_knowledge_boolean',
|
|
'nampdn-ai/tiny-textbooks',
|
|
'nampdn-ai/tiny-codes',
|
|
'bigcode/the-stack-smol-xs',
|
|
'm-a-p/CodeFeedback-Filtered-Instruction',
|
|
'jtatman/python-code-dataset-500k',
|
|
'iamtarun/python_code_instructions_18k_alpaca',
|
|
'HuggingFaceH4/CodeAlpaca_20K',
|
|
'gair-prox/open-web-math-pro',
|
|
'rvv-karma/Math-QA',
|
|
'ajibawa-2023/Maths-College',
|
|
'microsoft/orca-math-word-problems-200k',
|
|
'fblgit/simple-math',
|
|
'SkunkworksAI/reasoning-0.01',
|
|
'badrex/llm-emoji-dataset',
|
|
]
|
|
tags:
|
|
- litgpt
|
|
- litdata
|
|
---
|
|
|
|
# tangled-llama-58m-32k-base-v0.1
|
|
|
|
![logo](./misc/logo.png)
|
|
|
|
A pretrained language model based on the Llama model with about **58M** parameters. This model has been trained on **11.4B** (`11,422,750,857`) tokens from more than **0.8M** (`796,399`) dataset rows.
|
|
|
|
This model **isn't** designed for immediate use but rather for Continued Pretraining and Finetuning on a downstream task. While it can handle a context length of up to **128K** (`131,072`) tokens, it was pretrained with sequences of **2K** (`2048`) tokens.
|
|
|
|
The objective is to streamline the cognitive or reasoning core, eliminating any redundant knowledge from the model.
|
|
|
|
[loss, val_loss](https://api.wandb.ai/links/mtasic85/0i3wqsmb)
|
|
|
|
[val_ppl](https://api.wandb.ai/links/mtasic85/vz1jgu3v)
|
|
|
|
[epoch](https://api.wandb.ai/links/mtasic85/qltqthjr)
|
|
|
|
[learning_rate](https://api.wandb.ai/links/mtasic85/eqpiton4) |