models Llama 2 13b

Llama-2-13b

Overview

Meta has developed and publicly released the Llama 2 family of large language models (LLMs), a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama-2-Chat, are optimized for dialogue use cases. Llama-2-Chat models outperform open-source chat models on most benchmarks we tested, and in our human evaluations for helpfulness and safety, are on par with some popular closed-source models like ChatGPT and PaLM. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama-2-Chat in order to enable the community to build on our work and contribute to the responsible development of LLMs.

Note: Use of this model is governed by the Meta license. Click on View License above.

	Training Data	Params	Content Length	GQA	Tokens	LR
Llama 2	A new mix of publicly available online data	7B	4k	✗	2.0T	3.0 x 10^-4
Llama 2	A new mix of publicly available online data	13B	4k	✗	2.0T	3.0 x 10^-4
Llama 2	A new mix of publicly available online data	70B	4k	✔	2.0T	1.5 x 10^-4

Llama 2 family of models. Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger model -- 70B -- uses Grouped-Query Attention (GQA) for improved inference scalability.

Model Developers Meta AI

Variations Llama 2 comes in a range of parameter sizes — 7B, 13B, and 70B — as well as pretrained and fine-tuned variations.

Input Models input text only.

Output Models generate text only.

Model Architecture Llama 2 is an auto-regressive language optimized transformer. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.

Model Dates Llama 2 was trained between January 2023 and July 2023.

Status This is a static model trained on an offline dataset. Future versions of the tuned models will be released as we improve model safety with community feedback.

License A custom commercial license is available. Please see the Artifacts tab.

Where to send questions or comments about the model Instructions on how to provide feedback or comments on the model can be found in the model README, or by opening an issue in the GitHub repository.

Version: 24

Hardware and Software

Training Factors We used custom training libraries, Meta's Research Super Cluster, and production clusters for pretraining. Fine-tuning, annotation, and evaluation were also performed on third-party cloud compute.

Carbon Footprint Pretraining utilized a cumulative 3.3M GPU hours of computation on hardware of type A100-80GB (TDP of 350-400W). Estimated total emissions were 539 tCO2eq, 100% of which were offset by Meta’s sustainability program.

	Time (GPU hours)	Power Consumption (W)	Carbon Emitted(tCO₂eq)
Llama 2 7B	184320	400	31.22
Llama 2 13B	368640	400	62.44
Llama 2 70B	1720320	400	291.42
Total	3311616		539.00

CO₂ emissions during pretraining. Time: total GPU time required for training each model. Power Consumption: peak power capacity per GPU device for the GPUs used adjusted for power usage efficiency. 100% of the emissions are directly offset by Meta's sustainability program, and because we are openly releasing these models, the pretraining costs do not need to be incurred by others.

Training Data

Overview Llama 2 was pretrained on 2 trillion tokens of data from publicly available sources. The fine-tuning data includes publicly available instruction datasets, as well as over one million new human-annotated examples. Neither the pretraining nor the fine-tuning datasets include Meta user data.

Data Freshness The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023. The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.

Ethical Considerations and Limitations

Llama 2 is a new technology that carries risks with use. Testing conducted to date has not, and could not, cover all scenarios, including uses in languages other than English. For these reasons, as with all LLMs, Llama 2’s potential outputs cannot be predicted in advance, and the model may in some instances produce inaccurate, biased or other objectionable responses to user prompts. Therefore, before deploying any applications of Llama 2, developers should perform safety testing and tuning tailored to their specific applications of the model.

Please see the Responsible Use Guide available at https://ai.meta.com/llama/responsible-use-guide/

Finetuning samples

Task	Use case	Dataset	Python sample (Notebook)	CLI with YAML
Text Generation	Summarization	Samsum	summarization_with_text_gen.ipynb	text-generation.sh
Text Classification	Emotion Detection	Emotion	emotion-detection-llama.ipynb	emotion-detection.sh

Model Evaluation

Task	Use case	Dataset	Python sample (Notebook)	CLI with YAML
Text generation	Text generation	cnn_dailymail	evaluate-model-text-generation.ipynb	evaluate-model-text-generation.yml

Sample inputs and outputs (for real-time inference)

Supported Parameters

temperature: Controls randomness in the model. Lower values will make the model more deterministic and higher values will make the model more random.
max_new_tokens: The maximum number of tokens to generate.
top_k: The number of highest probability vocabulary tokens to keep for top-k-filtering. Default value is null, which disables top-k-filtering.
top_p: The cumulative probability of parameter highest probability vocabulary tokens to keep for nucleus sampling, defaults to null.
do_sample: Whether or not to use sampling; use greedy decoding otherwise.
return_full_text: Whether or not to return the full text (prompt + response) or only the generated part (response). Default value is false.
ignore_eos: Whether to ignore the EOS token and continue generating tokens after the EOS token is generated. Defaults to False.

List may not be complete.

Sample input

{ 
  "input_data": {
        "input_string": ["I believe the meaning of life is"],
        "parameters": {
                "top_p": 0.9,
                "temperature": 0.7,
                "max_new_tokens": 50,
                "do_sample": true
            }
    }
}

Sample output

[
    {
        "0": "I believe the meaning of life is to live.\nI think the meaning of life is to find your purpose and live it.\nThe meaning of life is to be happy.\n"
    }
]

evaluation : In this section, we report the results for the Llama 1 and Llama 2 models on standard academic benchmarks. For all the evaluations, we use our internal evaluations library. For these models, we always pick the best score between our evaluation framework and any publicly reported results.

Model	Size	Code	Commonsense Reasoning	World Knowledge	Reading Comprehension	Math	MMLU	BBH	AGI Eval
Llama 1	7B	14.1	60.8	46.2	58.5	6.95	35.1	30.3	23.9
Llama 1	13B	18.9	66.1	52.6	62.3	10.9	46.9	37.0	33.9
Llama 1	33B	26.0	70.0	58.4	67.6	21.4	57.8	39.8	41.7
Llama 1	65B	30.7	70.7	60.5	68.6	30.8	63.4	43.5	47.6
Llama 2	7B	16.8	63.9	48.9	61.3	14.6	45.3	32.6	29.3
Llama 2	13B	24.5	66.9	55.4	65.8	28.7	54.8	39.4	39.1
Llama 2	70B	37.5	71.9	63.6	69.4	35.2	68.9	51.2	54.2

Overall performance on grouped academic benchmarks. Code: We report the average pass@1 scores of our models on HumanEval and MBPP. Commonsense Reasoning: We report the average of PIQA, SIQA, HellaSwag, WinoGrande, ARC easy and challenge, OpenBookQA, and CommonsenseQA. We report 7-shot results for CommonSenseQA and 0-shot results for all other benchmarks. World Knowledge: We evaluate the 5-shot performance on NaturalQuestions and TriviaQA and report the average. Reading Comprehension: For reading comprehension, we report the 0-shot average on SQuAD, QuAC, and BoolQ. MATH: We report the average of the GSM8K (8 shot) and MATH (4 shot) benchmarks at top 1. Popular Aggregated Benchmarks: We report the overall results for MMLU (5 shot), Big Bench Hard (BBH) (3 shot), and AGI Eval (3–5 shot). For AGI Eval, we only evaluate on the English tasks and report the average.

		TruthfulQA	Toxigen
Llama 1	7B	27.42	23.00
Llama 1	13B	41.74	23.08
Llama 1	33B	44.19	22.57
Llama 1	65B	48.71	21.77
Llama 2	7B	33.29	21.25
Llama 2	13B	41.86	26.10
Llama 2	70B	50.18	24.60

Evaluation of pretrained LLMs on automatic safety benchmarks. For TruthfulQA, we present the percentage of generations that are both truthful and informative (the higher the better). For ToxiGen, we present the percentage of toxic generations (the smaller the better).

		TruthfulQA	Toxigen
Llama-2-Chat	7B	57.04	0.00
Llama-2-Chat	13B	62.18	0.00
Llama-2-Chat	70B	64.14	0.01

Evaluation of fine-tuned LLMs on different safety datasets. For TruthfulQA, we present the percentage of generations that are both truthful and informative (the higher the better). For ToxiGen, we present the percentage of toxic generations (the smaller the better). `

View in Studio: https://ml.azure.com/registries/azureml/models/Llama-2-13b/version/24

License: custom

Properties

languages: EN

evaluation-min-sku-spec: 12|2|224|1474

inference-min-sku-spec: 24|4|448|2900

finetune-min-sku-spec: 40|2|440|128

evaluation-recommended-sku: Standard_NC12s_v3, Standard_NC24s_v3, Standard_ND40rs_v2, Standard_ND96asr_v4, Standard_ND96amsr_A100_v4

inference-recommended-sku: Standard_NC24s_v3, Standard_ND40rs_v2, Standard_ND96asr_v4, Standard_ND96amsr_A100_v4

finetune-recommended-sku: Standard_ND40rs_v2, Standard_NC48ads_A100_v4, Standard_NC96ads_A100_v4, Standard_ND96asr_v4, Standard_ND96amsr_A100_v4

finetuning-tasks: text-classification,text-generation

azureml.copyRegistryFilesToWorkspace: True

SharedComputeCapacityEnabled: True

baseModelWeightsVersion: 1.0

baseWeightsId: Llama-2-13b

Wiki menu

Home
Reference Documentation
- Components
- Data
- Environments
- Models
Contributing

Provide feedback

Saved searches

Use saved searches to filter your results more quickly