base_model: TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T
datasets:
- cerebras/SlimPajama-627B
- bigcode/starcoderdata
inference: false
language:
- en
license: apache-2.0
model_creator: TinyLlama
model_name: TinyLlama-1.1B-intermediate-step-955k-token-2T
pipeline_tag: text-generation
quantized_by: afrideva
tags:
- gguf
- ggml
- quantized
- q2_k
- q3_k_m
- q4_k_m
- q5_k_m
- q6_k
- q8_0
TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T-GGUF
Quantized GGUF model files for TinyLlama-1.1B-intermediate-step-955k-token-2T from TinyLlama
| Name | Quant method | Size |
|---|---|---|
| tinyllama-1.1b-intermediate-step-955k-token-2t.q2_k.gguf | q2_k | None |
| tinyllama-1.1b-intermediate-step-955k-token-2t.q3_k_m.gguf | q3_k_m | None |
| tinyllama-1.1b-intermediate-step-955k-token-2t.q4_k_m.gguf | q4_k_m | None |
| tinyllama-1.1b-intermediate-step-955k-token-2t.q5_k_m.gguf | q5_k_m | None |
| tinyllama-1.1b-intermediate-step-955k-token-2t.q6_k.gguf | q6_k | None |
| tinyllama-1.1b-intermediate-step-955k-token-2t.q8_0.gguf | q8_0 | None |
Original Model Card:
https://github.com/jzhang38/TinyLlama
The TinyLlama project aims to pretrain a 1.1B Llama model on 3 trillion tokens. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01.
We adopted exactly the same architecture and tokenizer as Llama 2. This means TinyLlama can be plugged and played in many open-source projects built upon Llama. Besides, TinyLlama is compact with only 1.1B parameters. This compactness allows it to cater to a multitude of applications demanding a restricted computation and memory footprint.
This Model
This is an intermediate checkpoint with 995K steps and 2003B tokens.
Releases Schedule
We will be rolling out intermediate checkpoints following the below schedule. We also include some baseline models for comparison.
| Date | HF Checkpoint | Tokens | Step | HellaSwag Acc_norm |
|---|---|---|---|---|
| Baseline | StableLM-Alpha-3B | 800B | -- | 38.31 |
| Baseline | Pythia-1B-intermediate-step-50k-105b | 105B | 50k | 42.04 |
| Baseline | Pythia-1B | 300B | 143k | 47.16 |
| 2023-09-04 | TinyLlama-1.1B-intermediate-step-50k-105b | 105B | 50k | 43.50 |
| 2023-09-16 | -- | 500B | -- | -- |
| 2023-10-01 | -- | 1T | -- | -- |
| 2023-10-16 | -- | 1.5T | -- | -- |
| 2023-10-31 | -- | 2T | -- | -- |
| 2023-11-15 | -- | 2.5T | -- | -- |
| 2023-12-01 | -- | 3T | -- | -- |
How to use
You will need the transformers>=4.31 Do check the TinyLlama github page for more information.
from transformers import AutoTokenizer
import transformers
import torch
model = "TinyLlama/TinyLlama-1.1B-intermediate-step-955k-token-2T"
tokenizer = AutoTokenizer.from_pretrained(model)
pipeline = transformers.pipeline(
"text-generation",
model=model,
torch_dtype=torch.float16,
device_map="auto",
)
sequences = pipeline(
'The TinyLlama project aims to pretrain a 1.1B Llama model on 3 trillion tokens. With some proper optimization, we can achieve this within a span of "just" 90 days using 16 A100-40G GPUs 🚀🚀. The training has started on 2023-09-01.',
do_sample=True,
top_k=10,
num_return_sequences=1,
repetition_penalty=1.5,
eos_token_id=tokenizer.eos_token_id,
max_length=500,
)
for seq in sequences:
print(f"Result: {seq['generated_text']}")