From 8474b9797cfd6518a8232d23f69059d063c14f58 Mon Sep 17 00:00:00 2001 From: www-data Date: Sat, 28 Sep 2024 04:13:20 +0000 Subject: [PATCH 1/5] add assets identified by bot --- assets/tokyo.yaml | 28 ++++++++++++++++++++++++++++ 1 file changed, 28 insertions(+) diff --git a/assets/tokyo.yaml b/assets/tokyo.yaml index 6084473a..63ccea00 100644 --- a/assets/tokyo.yaml +++ b/assets/tokyo.yaml @@ -26,3 +26,31 @@ prohibited_uses: '' monitoring: unknown feedback: none +- type: model + name: Reflection Llama-3.1 70B + organization: Unknown + description: Reflection Llama-3.1 70B is an open-source LLM, trained with a new technique called Reflection-Tuning that teaches a LLM to detect mistakes in its reasoning and correct course. The model was trained on synthetic data generated by Glaive. + created_date: 2024-09-28 + url: https://huggingface.co/mattshumer/Reflection-70B + model_card: https://huggingface.co/mattshumer/Reflection-70B + modality: + explanation: "you can sample from Reflection Llama-3.1 70B using the same code, pipelines, etc. as any other Llama model." This implies that the model takes text input and produces text output. + value: Text; text + analysis: Unknown + size: + explanation: The size is in the model's name: "Reflection Llama-3.1 70B." + value: 70B parameters + dependencies: [Glaive, Llama 3.1] + training_emissions: Unknown + training_time: Unknown + training_hardware: Unknown + quality_control: The model uses a Reflection-Tuning technique which allows it to notice mistakes in its reasoning and correct them. + access: + explanation: "Reflection Llama-3.1 70B is an open-source LLM." + value: open + license: Unknown + intended_uses: The model is intended for complex reasoning and reflection tasks. It is designed to separate its internal thoughts and reasoning from its final answer. + prohibited_uses: Unknown + monitoring: Unknown + feedback: Unknown + From 9ad5f56f5e4a91a962564c22fee0bf4b34a03924 Mon Sep 17 00:00:00 2001 From: Rishi Bommasani <47439426+rishibommasani@users.noreply.github.com> Date: Thu, 16 Jan 2025 17:18:10 -0800 Subject: [PATCH 2/5] Update assets/tokyo.yaml --- assets/tokyo.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/assets/tokyo.yaml b/assets/tokyo.yaml index 63ccea00..68582064 100644 --- a/assets/tokyo.yaml +++ b/assets/tokyo.yaml @@ -35,7 +35,7 @@ model_card: https://huggingface.co/mattshumer/Reflection-70B modality: explanation: "you can sample from Reflection Llama-3.1 70B using the same code, pipelines, etc. as any other Llama model." This implies that the model takes text input and produces text output. - value: Text; text + value: text; text analysis: Unknown size: explanation: The size is in the model's name: "Reflection Llama-3.1 70B." From f66a2ac18cd55a66579cc31de3c7bd4bc5e8e50d Mon Sep 17 00:00:00 2001 From: Rishi Bommasani <47439426+rishibommasani@users.noreply.github.com> Date: Thu, 16 Jan 2025 17:18:15 -0800 Subject: [PATCH 3/5] Update assets/tokyo.yaml --- assets/tokyo.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/assets/tokyo.yaml b/assets/tokyo.yaml index 68582064..9a39039b 100644 --- a/assets/tokyo.yaml +++ b/assets/tokyo.yaml @@ -48,7 +48,7 @@ access: explanation: "Reflection Llama-3.1 70B is an open-source LLM." value: open - license: Unknown + license: Llama 3.1 Community License Agreement intended_uses: The model is intended for complex reasoning and reflection tasks. It is designed to separate its internal thoughts and reasoning from its final answer. prohibited_uses: Unknown monitoring: Unknown From 33301d918f46e217468d5e51af23da3c0dc4d6d9 Mon Sep 17 00:00:00 2001 From: Jonathan Xue <105090474+jxue16@users.noreply.github.com> Date: Thu, 16 Jan 2025 17:24:25 -0800 Subject: [PATCH 4/5] Update assets/tokyo.yaml --- assets/tokyo.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/assets/tokyo.yaml b/assets/tokyo.yaml index 8f43da0e..70fbd6ca 100644 --- a/assets/tokyo.yaml +++ b/assets/tokyo.yaml @@ -34,7 +34,7 @@ url: https://huggingface.co/mattshumer/Reflection-70B model_card: https://huggingface.co/mattshumer/Reflection-70B modality: - explanation: "you can sample from Reflection Llama-3.1 70B using the same code, pipelines, etc. as any other Llama model." This implies that the model takes text input and produces text output. + explanation: you can sample from Reflection Llama-3.1 70B using the same code, pipelines, etc. as any other Llama model. This implies that the model takes text input and produces text output. value: text; text analysis: Unknown size: From 901d91cec72ea6576e8b245450e0da7dec558d86 Mon Sep 17 00:00:00 2001 From: Jonathan Xue <105090474+jxue16@users.noreply.github.com> Date: Thu, 16 Jan 2025 17:24:41 -0800 Subject: [PATCH 5/5] Update assets/tokyo.yaml --- assets/tokyo.yaml | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/assets/tokyo.yaml b/assets/tokyo.yaml index 70fbd6ca..8ce5e078 100644 --- a/assets/tokyo.yaml +++ b/assets/tokyo.yaml @@ -38,7 +38,7 @@ value: text; text analysis: Unknown size: - explanation: The size is in the model's name: "Reflection Llama-3.1 70B." + explanation: The size is in the model's name, Reflection Llama-3.1 70B. value: 70B parameters dependencies: [Glaive, Llama 3.1] training_emissions: Unknown