EDGE-AI
/

EDGE_0-7B_GGML

@@ -4,12 +4,10 @@ language:
 - en
 pipeline_tag: text-generation
 tags:
-- facebook
-- meta
 - pytorch
-- llama
-- llama-2
-model_type: llama
 license: other
 ---
@@ -17,17 +15,17 @@ license: other
 <!-- header end -->
-# Meta's Llama 2 7B GGML
-These files are GGML format model files for [Meta's Llama 2 7B](https://huggingface.co/meta-llama/Llama-2-7b-hf).
-GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
 * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a powerful GGML web UI with full GPU acceleration out of the box. Especially good for story telling.
 * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with GPU acceleration via the c_transformers backend.
 * [LM Studio](https://lmstudio.ai/), a fully featured local GUI. Supports full GPU accel on macOS. Also supports Windows, without GPU accel.
-* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most popular web UI. Requires extra steps to enable GPU accel via llama.cpp backend.
 * [ctransformers](https://github.com/marella/ctransformers), a Python library with LangChain support and OpenAI-compatible AI server.
-* [llama-cpp-python](https://github.com/abetlen/llama-cpp-python), a Python library with OpenAI-compatible API server
 *
 ## Prompt template: None
@@ -39,15 +37,15 @@ GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/gger
 <!-- compatibility_ggml start -->
 ## Compatibility
-### Original llama.cpp quant methods: `q4_0, q4_1, q5_0, q5_1, q8_0`
 These are guaranteed to be compatible with any UIs, tools and libraries released since late May. They may be phased out soon, as they are largely superseded by the new k-quant methods.
 ### New k-quant methods: `q2_K, q3_K_S, q3_K_M, q3_K_L, q4_K_S, q4_K_M, q5_K_S, q6_K`
-These new quantisation methods are compatible with llama.cpp as of June 6th, commit `2d43387`.
-They are now also compatible with recent releases of text-generation-webui, KoboldCpp, llama-cpp-python, ctransformers, rustformers and most others. For compatibility with other tools and libraries, please check their documentation.
 ## Explanation of the new k-quant methods
 <details>
@@ -68,29 +66,29 @@ Refer to the Provided Files table below to see what files use which methods, and
 ## Provided files
 | Name | Quant method | Bits | Size | Max RAM required | Use case |
 | ---- | ---- | ---- | ---- | ---- | ----- |
-| llama-2-7b.ggmlv3.q2_K.bin | q2_K | 2 | 2.87 GB| 5.37 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors. |
-| llama-2-7b.ggmlv3.q3_K_L.bin | q3_K_L | 3 | 3.60 GB| 6.10 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
-| llama-2-7b.ggmlv3.q3_K_M.bin | q3_K_M | 3 | 3.28 GB| 5.78 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
-| llama-2-7b.ggmlv3.q3_K_S.bin | q3_K_S | 3 | 2.95 GB| 5.45 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
-| llama-2-7b.ggmlv3.q4_0.bin | q4_0 | 4 | 3.79 GB| 6.29 GB | Original quant method, 4-bit. |
-| llama-2-7b.ggmlv3.q4_1.bin | q4_1 | 4 | 4.21 GB| 6.71 GB | Original quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
-| llama-2-7b.ggmlv3.q4_K_M.bin | q4_K_M | 4 | 4.08 GB| 6.58 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
-| llama-2-7b.ggmlv3.q4_K_S.bin | q4_K_S | 4 | 3.83 GB| 6.33 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
-| llama-2-7b.ggmlv3.q5_0.bin | q5_0 | 5 | 4.63 GB| 7.13 GB | Original quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
-| llama-2-7b.ggmlv3.q5_1.bin | q5_1 | 5 | 5.06 GB| 7.56 GB | Original quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
-| llama-2-7b.ggmlv3.q5_K_M.bin | q5_K_M | 5 | 4.78 GB| 7.28 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
-| llama-2-7b.ggmlv3.q5_K_S.bin | q5_K_S | 5 | 4.65 GB| 7.15 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
-| llama-2-7b.ggmlv3.q6_K.bin | q6_K | 6 | 5.53 GB| 8.03 GB | New k-quant method. Uses GGML_TYPE_Q8_K for all tensors - 6-bit quantization |
-| llama-2-7b.ggmlv3.q8_0.bin | q8_0 | 8 | 7.16 GB| 9.66 GB | Original quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. |
 **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
-## How to run in `llama.cpp`
 I use the following command line; adjust for your tastes and needs:
 ```
-./main -t 10 -ngl 32 -m llama-2-7b.ggmlv3.q4_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### Instruction: Write a story about llamas\n### Response:"
 ```
 Change `-t 10` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`.
@@ -100,142 +98,142 @@ If you want to have a chat-style conversation, replace the `-p <PROMPT>` argumen
 ## How to run in `text-generation-webui`
-Further instructions here: [text-generation-webui/docs/llama.cpp-models.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/llama.cpp-models.md).
-# Original model card: Meta's Llama 2 7B
 ---
-extra_gated_heading: Access Llama 2 on Hugging Face
 extra_gated_description: >-
-  This is a form to enable access to Llama 2 on Hugging Face after you have been
-  granted access from Meta. Please visit the [Meta website](https://ai.meta.com/resources/models-and-libraries/llama-downloads) and accept our
   license terms and acceptable use policy before submitting this form. Requests
   will be processed in 1-2 days.
 extra_gated_button_content: Submit
 extra_gated_fields:
-  I agree to share my name, email address and username with Meta and confirm that I have already been granted download access on the Meta website: checkbox
 language:
 - en
 pipeline_tag: text-generation
 inference: false
 tags:
 - facebook
-- meta
 - pytorch
-- llama
-- llama-2
 ---
-# **Llama 2**
-Llama 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 7B pretrained model, converted for the Hugging Face Transformers format. Links to other models can be found in the index at the bottom.
 ## Model Details
-*Note: Use of this model is governed by the Meta license. In order to download the model weights and tokenizer, please visit the [website](https://ai.meta.com/resources/models-and-libraries/llama-downloads/) and accept our License before requesting access here.*
-Meta 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.
-**Model Developers** Meta
-**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 model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
 ||Training Data|Params|Content Length|GQA|Tokens|LR|
 |---|---|---|---|---|---|---|
-|Llama 2|*A new mix of publicly available online data*|7B|4k|&#10007;|2.0T|3.0 x 10<sup>-4</sup>|
-|Llama 2|*A new mix of publicly available online data*|13B|4k|&#10007;|2.0T|3.0 x 10<sup>-4</sup>|
-|Llama 2|*A new mix of publicly available online data*|70B|4k|&#10004;|2.0T|1.5 x 10<sup>-4</sup>|
-*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 models -  70B -- use Grouped-Query Attention (GQA) for improved inference scalability.
-**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 at: [https://ai.meta.com/resources/models-and-libraries/llama-downloads/](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
 ## Intended Use
-**Intended Use Cases** Llama 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
-To get the expected features and performance for the chat versions, a specific formatting needs to be followed, including the `INST` and `<<SYS>>` tags, `BOS` and `EOS` tokens, and the whitespaces and breaklines in between (we recommend calling `strip()` on inputs to avoid double-spaces). See our reference code in github for details: [`chat_completion`](https://github.com/facebookresearch/llama/blob/main/llama/generation.py#L212).
-**Out-of-scope Uses** Use in any manner that violates applicable laws or regulations (including trade compliance laws).Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for Llama 2.
 ## 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(tCO<sub>2</sub>eq)|
 |---|---|---|---|
-|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|
-**CO<sub>2</sub> 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.
 ## Evaluation Results
-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.
 |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.
 |||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.** Same metric definitions as above.
 ## Ethical Considerations and Limitations
-Llama 2 is a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. 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/](https://ai.meta.com/llama/responsible-use-guide)
 ## Reporting Issues
 Please report any software “bug,” or other problems with the models through one of the following means:
-- Reporting issues with the model: [github.com/facebookresearch/llama](http://github.com/facebookresearch/llama)
-- Reporting problematic content generated by the model: [developers.facebook.com/llama_output_feedback](http://developers.facebook.com/llama_output_feedback)
 - Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)
-## Llama Model Index
-|Model|Llama2|Llama2-hf|Llama2-chat|Llama2-chat-hf|
 |---|---|---|---|---|
-|7B| [Link](https://huggingface.co/llamaste/Llama-2-7b) | [Link](https://huggingface.co/llamaste/Llama-2-7b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-7b-chat-hf)|
-|13B| [Link](https://huggingface.co/llamaste/Llama-2-13b) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-13b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-13b-hf)|
-|70B| [Link](https://huggingface.co/llamaste/Llama-2-70b) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf) | [Link](https://huggingface.co/llamaste/Llama-2-70b-chat) | [Link](https://huggingface.co/llamaste/Llama-2-70b-hf)|

 - en
 pipeline_tag: text-generation
 tags:
 - pytorch
+- EDGE
+- EDGE-2
+model_type: EDGE
 license: other
 ---
 <!-- header end -->
+# EDGE 2 7B GGML
+These files are GGML format model files for ['s EDGE 2 7B](https://huggingface.co/-EDGE/EDGE-2-7b-hf).
+GGML files are for CPU + GPU inference using [EDGE.cpp](https://github.com/ggerganov/EDGE.cpp) and libraries and UIs which support this format, such as:
 * [KoboldCpp](https://github.com/LostRuins/koboldcpp), a powerful GGML web UI with full GPU acceleration out of the box. Especially good for story telling.
 * [LoLLMS Web UI](https://github.com/ParisNeo/lollms-webui), a great web UI with GPU acceleration via the c_transformers backend.
 * [LM Studio](https://lmstudio.ai/), a fully featured local GUI. Supports full GPU accel on macOS. Also supports Windows, without GPU accel.
+* [text-generation-webui](https://github.com/oobabooga/text-generation-webui), the most popular web UI. Requires extra steps to enable GPU accel via EDGE.cpp backend.
 * [ctransformers](https://github.com/marella/ctransformers), a Python library with LangChain support and OpenAI-compatible AI server.
+* [EDGE-cpp-python](https://github.com/abetlen/EDGE-cpp-python), a Python library with OpenAI-compatible API server
 *
 ## Prompt template: None
 <!-- compatibility_ggml start -->
 ## Compatibility
+### Original EDGE.cpp quant methods: `q4_0, q4_1, q5_0, q5_1, q8_0`
 These are guaranteed to be compatible with any UIs, tools and libraries released since late May. They may be phased out soon, as they are largely superseded by the new k-quant methods.
 ### New k-quant methods: `q2_K, q3_K_S, q3_K_M, q3_K_L, q4_K_S, q4_K_M, q5_K_S, q6_K`
+These new quantisation methods are compatible with EDGE.cpp as of June 6th, commit `2d43387`.
+They are now also compatible with recent releases of text-generation-webui, KoboldCpp, EDGE-cpp-python, ctransformers, rustformers and most others. For compatibility with other tools and libraries, please check their documentation.
 ## Explanation of the new k-quant methods
 <details>
 ## Provided files
 | Name | Quant method | Bits | Size | Max RAM required | Use case |
 | ---- | ---- | ---- | ---- | ---- | ----- |
+| EDGE-2-7b.ggmlv3.q2_K.bin | q2_K | 2 | 2.87 GB| 5.37 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.vw and feed_forward.w2 tensors, GGML_TYPE_Q2_K for the other tensors. |
+| EDGE-2-7b.ggmlv3.q3_K_L.bin | q3_K_L | 3 | 3.60 GB| 6.10 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
+| EDGE-2-7b.ggmlv3.q3_K_M.bin | q3_K_M | 3 | 3.28 GB| 5.78 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
+| EDGE-2-7b.ggmlv3.q3_K_S.bin | q3_K_S | 3 | 2.95 GB| 5.45 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
+| EDGE-2-7b.ggmlv3.q4_0.bin | q4_0 | 4 | 3.79 GB| 6.29 GB | Original quant method, 4-bit. |
+| EDGE-2-7b.ggmlv3.q4_1.bin | q4_1 | 4 | 4.21 GB| 6.71 GB | Original quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
+| EDGE-2-7b.ggmlv3.q4_K_M.bin | q4_K_M | 4 | 4.08 GB| 6.58 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
+| EDGE-2-7b.ggmlv3.q4_K_S.bin | q4_K_S | 4 | 3.83 GB| 6.33 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
+| EDGE-2-7b.ggmlv3.q5_0.bin | q5_0 | 5 | 4.63 GB| 7.13 GB | Original quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
+| EDGE-2-7b.ggmlv3.q5_1.bin | q5_1 | 5 | 5.06 GB| 7.56 GB | Original quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
+| EDGE-2-7b.ggmlv3.q5_K_M.bin | q5_K_M | 5 | 4.78 GB| 7.28 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
+| EDGE-2-7b.ggmlv3.q5_K_S.bin | q5_K_S | 5 | 4.65 GB| 7.15 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
+| EDGE-2-7b.ggmlv3.q6_K.bin | q6_K | 6 | 5.53 GB| 8.03 GB | New k-quant method. Uses GGML_TYPE_Q8_K for all tensors - 6-bit quantization |
+| EDGE-2-7b.ggmlv3.q8_0.bin | q8_0 | 8 | 7.16 GB| 9.66 GB | Original quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. |
 **Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
+## How to run in `EDGE.cpp`
 I use the following command line; adjust for your tastes and needs:
 ```
+./main -t 10 -ngl 32 -m EDGE-2-7b.ggmlv3.q4_0.bin --color -c 2048 --temp 0.7 --repeat_penalty 1.1 -n -1 -p "### Instruction: Write a story about EDGEs\n### Response:"
 ```
 Change `-t 10` to the number of physical CPU cores you have. For example if your system has 8 cores/16 threads, use `-t 8`.
 ## How to run in `text-generation-webui`
+Further instructions here: [text-generation-webui/docs/EDGE.cpp-models.md](https://github.com/oobabooga/text-generation-webui/blob/main/docs/EDGE.cpp-models.md).
+# Original model card: 's EDGE 2 7B
 ---
+extra_gated_heading: Access EDGE 2 on Hugging Face
 extra_gated_description: >-
+  This is a form to enable access to EDGE 2 on Hugging Face after you have been
+  granted access from . Please visit the [ website](https://ai..com/resources/models-and-libraries/EDGE-downloads) and accept our
   license terms and acceptable use policy before submitting this form. Requests
   will be processed in 1-2 days.
 extra_gated_button_content: Submit
 extra_gated_fields:
+  I agree to share my name, email address and username with  and confirm that I have already been granted download access on the  website: checkbox
 language:
 - en
 pipeline_tag: text-generation
 inference: false
 tags:
 - facebook
+-
 - pytorch
+- EDGE
+- EDGE-2
 ---
+# **EDGE 2**
+EDGE 2 is a collection of pretrained and fine-tuned generative text models ranging in scale from 7 billion to 70 billion parameters. This is the repository for the 7B pretrained model, converted for the Hugging Face Transformers format. Links to other models can be found in the index at the bottom.
 ## Model Details
+*Note: Use of this model is governed by the  license. In order to download the model weights and tokenizer, please visit the [website](https://ai..com/resources/models-and-libraries/EDGE-downloads/) and accept our License before requesting access here.*
+ developed and publicly released the EDGE 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 EDGE-2-Chat, are optimized for dialogue use cases. EDGE-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.
+**Model Developers**
+**Variations** EDGE 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** EDGE 2 is an auto-regressive language model that uses an optimized transformer architecture. The tuned versions use supervised fine-tuning (SFT) and reinforcement learning with human feedback (RLHF) to align to human preferences for helpfulness and safety.
 ||Training Data|Params|Content Length|GQA|Tokens|LR|
 |---|---|---|---|---|---|---|
+|EDGE 2|*A new mix of publicly available online data*|7B|4k|&#10007;|2.0T|3.0 x 10<sup>-4</sup>|
+|EDGE 2|*A new mix of publicly available online data*|13B|4k|&#10007;|2.0T|3.0 x 10<sup>-4</sup>|
+|EDGE 2|*A new mix of publicly available online data*|70B|4k|&#10004;|2.0T|1.5 x 10<sup>-4</sup>|
+*EDGE 2 family of models.* Token counts refer to pretraining data only. All models are trained with a global batch-size of 4M tokens. Bigger models -  70B -- use Grouped-Query Attention (GQA) for improved inference scalability.
+**Model Dates** EDGE 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 at: [https://ai..com/resources/models-and-libraries/EDGE-downloads/](https://ai..com/resources/models-and-libraries/EDGE-downloads/)
 ## Intended Use
+**Intended Use Cases** EDGE 2 is intended for commercial and research use in English. Tuned models are intended for assistant-like chat, whereas pretrained models can be adapted for a variety of natural language generation tasks.
+To get the expected features and performance for the chat versions, a specific formatting needs to be followed, including the `INST` and `<<SYS>>` tags, `BOS` and `EOS` tokens, and the whitespaces and breaklines in between (we recommend calling `strip()` on inputs to avoid double-spaces). See our reference code in github for details: [`chat_completion`](https://github.com/facebookresearch/EDGE/blob/main/EDGE/generation.py#L212).
+**Out-of-scope Uses** Use in any manner that violates applicable laws or regulations (including trade compliance laws).Use in languages other than English. Use in any other way that is prohibited by the Acceptable Use Policy and Licensing Agreement for EDGE 2.
 ## Hardware and Software
+**Training Factors** We used custom training libraries, '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 ’s sustainability program.
 ||Time (GPU hours)|Power Consumption (W)|Carbon Emitted(tCO<sub>2</sub>eq)|
 |---|---|---|---|
+|EDGE 2 7B|184320|400|31.22|
+|EDGE 2 13B|368640|400|62.44|
+|EDGE 2 70B|1720320|400|291.42|
 |Total|3311616||539.00|
+**CO<sub>2</sub> 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 '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** EDGE 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  user data.
 **Data Freshness** The pretraining data has a cutoff of September 2022, but some tuning data is more recent, up to July 2023.
 ## Evaluation Results
+In this section, we report the results for the EDGE 1 and EDGE 2 models on standard academic benchmarks.For all the evaluations, we use our internal evaluations library.
 |Model|Size|Code|Commonsense Reasoning|World Knowledge|Reading Comprehension|Math|MMLU|BBH|AGI Eval|
 |---|---|---|---|---|---|---|---|---|---|
+|EDGE 1|7B|14.1|60.8|46.2|58.5|6.95|35.1|30.3|23.9|
+|EDGE 1|13B|18.9|66.1|52.6|62.3|10.9|46.9|37.0|33.9|
+|EDGE 1|33B|26.0|70.0|58.4|67.6|21.4|57.8|39.8|41.7|
+|EDGE 1|65B|30.7|70.7|60.5|68.6|30.8|63.4|43.5|47.6|
+|EDGE 2|7B|16.8|63.9|48.9|61.3|14.6|45.3|32.6|29.3|
+|EDGE 2|13B|24.5|66.9|55.4|65.8|28.7|54.8|39.4|39.1|
+|EDGE 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.
 |||TruthfulQA|Toxigen|
 |---|---|---|---|
+|EDGE 1|7B|27.42|23.00|
+|EDGE 1|13B|41.74|23.08|
+|EDGE 1|33B|44.19|22.57|
+|EDGE 1|65B|48.71|21.77|
+|EDGE 2|7B|33.29|**21.25**|
+|EDGE 2|13B|41.86|26.10|
+|EDGE 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|
 |---|---|---|---|
+|EDGE-2-Chat|7B|57.04|**0.00**|
+|EDGE-2-Chat|13B|62.18|**0.00**|
+|EDGE-2-Chat|70B|**64.14**|0.01|
 **Evaluation of fine-tuned LLMs on different safety datasets.** Same metric definitions as above.
 ## Ethical Considerations and Limitations
+EDGE 2 is a new technology that carries risks with use. Testing conducted to date has been in English, and has not covered, nor could it cover all scenarios. For these reasons, as with all LLMs, EDGE 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 EDGE 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..com/EDGE/responsible-use-guide/](https://ai..com/EDGE/responsible-use-guide)
 ## Reporting Issues
 Please report any software “bug,” or other problems with the models through one of the following means:
+- Reporting issues with the model: [github.com/facebookresearch/EDGE](http://github.com/facebookresearch/EDGE)
+- Reporting problematic content generated by the model: [developers.facebook.com/EDGE_output_feedback](http://developers.facebook.com/EDGE_output_feedback)
 - Reporting bugs and security concerns: [facebook.com/whitehat/info](http://facebook.com/whitehat/info)
+## EDGE Model Index
+|Model|EDGE2|EDGE2-hf|EDGE2-chat|EDGE2-chat-hf|
 |---|---|---|---|---|
+|7B| [Link](https://huggingface.co/EDGEste/EDGE-2-7b) | [Link](https://huggingface.co/EDGEste/EDGE-2-7b-hf) | [Link](https://huggingface.co/EDGEste/EDGE-2-7b-chat) | [Link](https://huggingface.co/EDGEste/EDGE-2-7b-chat-hf)|
+|13B| [Link](https://huggingface.co/EDGEste/EDGE-2-13b) | [Link](https://huggingface.co/EDGEste/EDGE-2-13b-hf) | [Link](https://huggingface.co/EDGEste/EDGE-2-13b-chat) | [Link](https://huggingface.co/EDGEste/EDGE-2-13b-hf)|
+|70B| [Link](https://huggingface.co/EDGEste/EDGE-2-70b) | [Link](https://huggingface.co/EDGEste/EDGE-2-70b-hf) | [Link](https://huggingface.co/EDGEste/EDGE-2-70b-chat) | [Link](https://huggingface.co/EDGEste/EDGE-2-70b-hf)|