Upload folder using huggingface_hub

README.md

@@ -0,0 +1,142 @@
+---
+license: apache-2.0
+tags:
+- finetuned
+pipeline_tag: text-generation
+new_version: mistralai/Mistral-7B-Instruct-v0.3
+inference: true
+widget:
+- messages:
+  - role: user
+    content: What is your favorite condiment?
+
+extra_gated_description: If you want to learn more about how we process your personal data, please read our <a href="https://mistral.ai/terms/">Privacy Policy</a>.
+---
+
+# Model Card for Mistral-7B-Instruct-v0.2
+
+
+## Encode and Decode with `mistral_common`
+            
+```py
+from mistral_common.tokens.tokenizers.mistral import MistralTokenizer
+from mistral_common.protocol.instruct.messages import UserMessage
+from mistral_common.protocol.instruct.request import ChatCompletionRequest
+ 
+mistral_models_path = "MISTRAL_MODELS_PATH"
+ 
+tokenizer = MistralTokenizer.v1()
+ 
+completion_request = ChatCompletionRequest(messages=[UserMessage(content="Explain Machine Learning to me in a nutshell.")])
+ 
+tokens = tokenizer.encode_chat_completion(completion_request).tokens
+```
+ 
+## Inference with `mistral_inference`
+ 
+ ```py
+from mistral_inference.transformer import Transformer
+from mistral_inference.generate import generate
+ 
+model = Transformer.from_folder(mistral_models_path)
+out_tokens, _ = generate([tokens], model, max_tokens=64, temperature=0.0, eos_id=tokenizer.instruct_tokenizer.tokenizer.eos_id)
+
+result = tokenizer.decode(out_tokens[0])
+
+print(result)
+```
+
+## Inference with hugging face `transformers`
+ 
+```py
+from transformers import AutoModelForCausalLM
+ 
+model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
+model.to("cuda")
+ 
+generated_ids = model.generate(tokens, max_new_tokens=1000, do_sample=True)
+
+# decode with mistral tokenizer
+result = tokenizer.decode(generated_ids[0].tolist())
+print(result)
+```
+
+> [!TIP]
+> PRs to correct the `transformers` tokenizer so that it gives 1-to-1 the same results as the `mistral_common` reference implementation are very welcome!
+            
+---
+
+The Mistral-7B-Instruct-v0.2 Large Language Model (LLM) is an instruct fine-tuned version of the Mistral-7B-v0.2.
+
+Mistral-7B-v0.2 has the following changes compared to Mistral-7B-v0.1
+- 32k context window (vs 8k context in v0.1)
+- Rope-theta = 1e6
+- No Sliding-Window Attention
+
+For full details of this model please read our [paper](https://arxiv.org/abs/2310.06825) and [release blog post](https://mistral.ai/news/la-plateforme/).
+
+## Instruction format
+
+In order to leverage instruction fine-tuning, your prompt should be surrounded by `[INST]` and `[/INST]` tokens. The very first instruction should begin with a begin of sentence id. The next instructions should not. The assistant generation will be ended by the end-of-sentence token id.
+
+E.g.
+```
+text = "<s>[INST] What is your favourite condiment? [/INST]"
+"Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!</s> "
+"[INST] Do you have mayonnaise recipes? [/INST]"
+```
+
+This format is available as a [chat template](https://huggingface.co/docs/transformers/main/chat_templating) via the `apply_chat_template()` method:
+
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+device = "cuda" # the device to load the model onto
+
+model = AutoModelForCausalLM.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
+tokenizer = AutoTokenizer.from_pretrained("mistralai/Mistral-7B-Instruct-v0.2")
+
+messages = [
+    {"role": "user", "content": "What is your favourite condiment?"},
+    {"role": "assistant", "content": "Well, I'm quite partial to a good squeeze of fresh lemon juice. It adds just the right amount of zesty flavour to whatever I'm cooking up in the kitchen!"},
+    {"role": "user", "content": "Do you have mayonnaise recipes?"}
+]
+
+encodeds = tokenizer.apply_chat_template(messages, return_tensors="pt")
+
+model_inputs = encodeds.to(device)
+model.to(device)
+
+generated_ids = model.generate(model_inputs, max_new_tokens=1000, do_sample=True)
+decoded = tokenizer.batch_decode(generated_ids)
+print(decoded[0])
+```
+
+## Troubleshooting
+- If you see the following error:
+```
+Traceback (most recent call last):
+File "", line 1, in
+File "/transformers/models/auto/auto_factory.py", line 482, in from_pretrained
+config, kwargs = AutoConfig.from_pretrained(
+File "/transformers/models/auto/configuration_auto.py", line 1022, in from_pretrained
+config_class = CONFIG_MAPPING[config_dict["model_type"]]
+File "/transformers/models/auto/configuration_auto.py", line 723, in getitem
+raise KeyError(key)
+KeyError: 'mistral'
+```
+
+Installing transformers from source should solve the issue
+pip install git+https://github.com/huggingface/transformers
+
+This should not be required after transformers-v4.33.4.
+
+## Limitations
+
+The Mistral 7B Instruct model is a quick demonstration that the base model can be easily fine-tuned to achieve compelling performance. 
+It does not have any moderation mechanisms. We're looking forward to engaging with the community on ways to
+make the model finely respect guardrails, allowing for deployment in environments requiring moderated outputs.
+
+## The Mistral AI Team
+
+Albert Jiang, Alexandre Sablayrolles, Arthur Mensch, Blanche Savary, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, Gianna Lengyel, Guillaume Bour, Guillaume Lample, Lélio Renard Lavaud, Louis Ternon, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Théophile Gervet, Thibaut Lavril, Thomas Wang, Timothée Lacroix, William El Sayed.

config.json

@@ -0,0 +1,25 @@
+{
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 32768,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 1000000.0,
+  "sliding_window": null,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.36.0",
+  "use_cache": true,
+  "vocab_size": 32000
+}

ft-model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8a552a6521234a5ee1e5e31bc5b2c8745f6825124c464169bb6ff714b1c7eab
+size 4943162336

ft-model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b5736bc4ae253ca3bb265cf677f3aa59eca80e052219bd3aa6b1279243946f6d
+size 4999819336

ft-model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a90ff2622fcdb78ce19cedc256eceeab67f68835cdded945b05a6353044f08ce
+size 4540516344

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.36.0"
+}

model.safetensors.index.json

@@ -0,0 +1,298 @@
+{
+  "metadata": {
+    "total_size": 14483464192
+  },
+  "weight_map": {
+    "model.embed_tokens.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.0.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.1.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.2.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.3.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.4.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.5.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.6.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.7.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.8.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.mlp.down_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.input_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.9.post_attention_layernorm.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.q_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.k_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.v_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.self_attn.o_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.gate_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.up_proj.weight": "ft-model-00001-of-00003.safetensors",
+    "model.layers.10.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.10.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.10.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.11.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.12.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.13.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.14.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.15.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.16.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.17.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.18.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.19.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.20.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.gate_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.up_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.mlp.down_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.input_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.21.post_attention_layernorm.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.q_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.k_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.v_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.22.self_attn.o_proj.weight": "ft-model-00002-of-00003.safetensors",
+    "model.layers.22.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.22.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.22.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.22.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.23.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.24.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.25.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.26.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.27.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.28.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.29.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.30.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.q_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.k_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.v_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.self_attn.o_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.gate_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.up_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.mlp.down_proj.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.input_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.layers.31.post_attention_layernorm.weight": "ft-model-00003-of-00003.safetensors",
+    "model.norm.weight": "ft-model-00003-of-00003.safetensors",
+    "lm_head.weight": "ft-model-00003-of-00003.safetensors"
+  }
+}

original_repo_id.json

@@ -0,0 +1,3 @@
+{
+    "repo_id": "mistralai/Mistral-7B-Instruct-v0.2"
+}

special_tokens_map.json

@@ -0,0 +1,23 @@
+{
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "</s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "add_prefix_space": null,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [],
+  "bos_token": "<s>",
+  "chat_template": "{%- if messages[0]['role'] == 'system' %}\n    {%- set system_message = messages[0]['content'] %}\n    {%- set loop_messages = messages[1:] %}\n{%- else %}\n    {%- set loop_messages = messages %}\n{%- endif %}\n\n{{- bos_token }}\n{%- for message in loop_messages %}\n    {%- if (message['role'] == 'user') != (loop.index0 % 2 == 0) %}\n        {{- raise_exception('After the optional system message, conversation roles must alternate user/assistant/user/assistant/...') }}\n    {%- endif %}\n    {%- if message['role'] == 'user' %}\n        {%- if loop.first and system_message is defined %}\n            {{- ' [INST] ' + system_message + '\\n\\n' + message['content'] + ' [/INST]' }}\n        {%- else %}\n            {{- ' [INST] ' + message['content'] + ' [/INST]' }}\n        {%- endif %}\n    {%- elif message['role'] == 'assistant' %}\n        {{- ' ' + message['content'] + eos_token}}\n    {%- else %}\n        {{- raise_exception('Only user and assistant roles are supported, with the exception of an initial optional system message!') }}\n    {%- endif %}\n{%- endfor %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "</s>",
+  "legacy": false,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": null,
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": false
+}

torchtune_config.yaml

@@ -0,0 +1,110 @@
+prefix: /home/arqa39/.torchtune
+output_dir: /home/arqa39/.torchtune/fed_ppo/mistral_7b
+tokenizer:
+  _component_: torchtune.models.mistral.mistral_tokenizer
+  path: ${prefix}/models/Mistral-7B-Instruct-v0.2/tokenizer.model
+  max_seq_len: null
+dataset:
+  _component_: torchtune.datasets.text_completion_dataset
+  source: trl-internal-testing/sentiment-trl-style
+  split: train
+  column: prompt
+  add_eos: false
+policy:
+  _component_: torchtune.models.mistral.lora_mistral_7b
+  lora_attn_modules:
+  - q_proj
+  - k_proj
+  - v_proj
+  - output_proj
+  apply_lora_to_mlp: true
+  apply_lora_to_output: false
+  lora_rank: 64
+  lora_alpha: 16
+  lora_dropout: 0.0
+  quantize_base: false
+valmod:
+  _component_: torchtune.models.mistral._component_builders.lora_mistral_classifier
+  attn_dropout: 0.0
+  embed_dim: 4096
+  intermediate_dim: 14336
+  max_seq_len: 32768
+  norm_eps: 1.0e-05
+  num_classes: 1
+  num_heads: 32
+  num_kv_heads: 8
+  num_layers: 32
+  vocab_size: 32001
+  lora_attn_modules:
+  - q_proj
+  - k_proj
+  - v_proj
+  - output_proj
+  apply_lora_to_mlp: true
+  apply_lora_to_output: = True
+  lora_rank: 16
+  lora_alpha: 32
+  lora_dropout: 0.0
+  quantize_base: false
+checkpointer:
+  _component_: torchtune.training.FullModelHFCheckpointer
+  checkpoint_dir: ${prefix}/models/Mistral-7B-Instruct-v0.2/
+  checkpoint_files:
+  - pytorch_model-00001-of-00003.bin
+  - pytorch_model-00002-of-00003.bin
+  - pytorch_model-00003-of-00003.bin
+  recipe_checkpoint: null
+  output_dir: ${output_dir}/policy
+  model_type: MISTRAL
+value_checkpointer:
+  _component_: torchtune.training.FullModelHFCheckpointer
+  checkpoint_dir: ${prefix}/models/RM-Mistral-7B/
+  checkpoint_files:
+  - model-00001-of-00003.safetensors
+  - model-00002-of-00003.safetensors
+  - model-00003-of-00003.safetensors
+  output_dir: ${output_dir}/value
+  model_type: REWARD
+seed: 53710
+shuffle: true
+device: cuda
+batch_size: 64
+num_steps: 10000
+ppo_epochs: 2
+ppo_batch_size: 32
+gradient_accumulation_steps: 1
+compile: false
+optimizer:
+  _component_: bitsandbytes.optim.PagedAdamW
+  lr: 0.0001
+optimizer_in_bwd: true
+log_peak_memory_stats: true
+enable_activation_checkpointing: true
+dtype: bf16
+forward_batch_size: 16
+max_generated_tokens: 58
+temperature: 0.7
+top_k: null
+min_response_length: 18
+penalise_no_eos: true
+reward_penalty: -3
+stop_token_ids:
+- 2
+- 28723
+whiten_rewards: false
+gamma: 1
+lmbda: 0.95
+loss:
+  _component_: torchtune.rlhf.loss.PPOLoss
+  epsilon: 0.2
+  value_coeff: 0.1
+  value_clip_range: 0.2
+kl_coeff: 0.01
+wandb_logger:
+  dir: ${prefix}
+  entity: RADFAN
+  project: FedPPO
+  group: SelfReference
+  name: Mistral-7B-LoRA-SelfRef-U13
+log_every_n_steps: 1
+update_ref_policy_every_n_steps: 13