Upload 13 files

README.md

@@ -0,0 +1,58 @@
+---
+tags:
+- autotrain
+- text-generation
+base_model: ahxt/llama2_xs_460M_experimental
+datasets:
+- KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format
+widget:
+- text: |-
+    ### Instruction:
+    Find me a list of some nice places to visit around the world.
+    
+    ### Response:
+- text: |-
+    ### Instruction:
+    Tell me a story about some magical place.
+    
+    ### Response:
+- text: |-
+    ### Instruction:
+    Tell me all you know about the Earth.
+    
+    ### Response:
+inference:
+  parameters:
+    max_new_tokens: 32
+    repetition_penalty: 1.15
+    do_sample: true
+    temperature: 0.5
+    top_p: 0.5
+---
+
+# ahxt's llama2_xs_460M_experimental trained on the WizardLM's Evol Instruct dataset using AutoTrain
+
+- Base model: [ahxt/llama2_xs_460M_experimental](https://huggingface.co/ahxt/llama2_xs_460M_experimental)
+- Dataset: [KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format](https://huggingface.co/datasets/KnutJaegersberg/WizardLM_evol_instruct_V2_196k_instruct_format)
+- [Training hyperparameters](https://huggingface.co/Felladrin/llama2_xs_460M_experimental_evol_instruct/blob/cc151c5669ea37c3ef972e375c74f2d9bfd92b49/training_params.json)
+- Availability in other ML formats:
+  - GGUF: [afrideva/llama2_xs_460M_experimental_evol_instruct-GGUF](https://huggingface.co/afrideva/llama2_xs_460M_experimental_evol_instruct-GGUF)
+  - ONNX: [Felladrin/onnx-llama2_xs_460M_experimental_evol_instruct](https://huggingface.co/Felladrin/onnx-llama2_xs_460M_experimental_evol_instruct)
+
+## Recommended Prompt Format
+
+```
+### Instruction:
+<instruction>
+
+### Response:
+```
+
+## Recommended Inference Parameters
+
+```yml
+repetition_penalty: 1.15
+do_sample: true
+temperature: 0.5
+top_p: 0.5
+```

adapter_config.json

@@ -0,0 +1,23 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "ahxt/llama2_xs_460M_experimental",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 32,
+  "lora_dropout": 0.1,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "v_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:53b8589368e7d965a89fa5f0c5d7495fe055cb575b1c4e1a3bb20ee273e361a6
+size 4927904

config.json

@@ -0,0 +1,29 @@
+{
+  "_name_or_path": "ahxt/llama2_xs_460M_experimental",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "hidden_act": "silu",
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 4096,
+  "layernorm_epsilon": 1e-05,
+  "max_position_embeddings": 1024,
+  "model_type": "llama",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "pad_token_id": 0,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": null,
+  "rope_theta": 10000.0,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.35.2",
+  "use_cache": true,
+  "vocab_size": 50304
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "pad_token_id": 0,
+  "transformers_version": "4.35.2"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:faf6159a5b12ff06354bc3510cf483742f39185769b93c0eafd1269347f0b060
+size 923396200

special_tokens_map.json

@@ -0,0 +1,30 @@
+{
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "#",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "\"",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": true,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,38 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "1": {
+      "content": "\"",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "#",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50256": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "bos_token": "<|endoftext|>",
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "#",
+  "errors": "replace",
+  "model_max_length": 1024,
+  "pad_token": "\"",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f4703f7a46c6e7c46bf6292a43bd139a14c82f330229a708da5c88ea553aec5
+size 4600

training_params.json

@@ -0,0 +1 @@
+{"model": "ahxt/llama2_xs_460M_experimental", "data_path": "Felladrin/autotrain-data-llama2_xs_460M_experimental_instruct", "project_name": "/tmp/model", "train_split": "train", "valid_split": null, "text_column": "autotrain_text", "rejected_text_column": null, "lr": 0.0002, "epochs": 1, "batch_size": 2, "warmup_ratio": 0.1, "gradient_accumulation": 1, "optimizer": "adamw_torch", "scheduler": "linear", "weight_decay": 0.01, "max_grad_norm": 1.0, "seed": 42, "add_eos_token": true, "block_size": 1024, "use_peft": true, "lora_r": 16, "lora_alpha": 32, "lora_dropout": 0.1, "logging_steps": -1, "evaluation_strategy": "epoch", "save_total_limit": 1, "save_strategy": "epoch", "auto_find_batch_size": false, "fp16": true, "push_to_hub": true, "use_int8": false, "model_max_length": 2048, "repo_id": "Felladrin/llama2_xs_460M_experimental_instruct-0", "use_int4": true, "trainer": "sft", "target_modules": null, "merge_adapter": false, "username": "Felladrin", "use_flash_attention_2": false, "log": "none", "disable_gradient_checkpointing": false, "model_ref": null, "dpo_beta": 0.1, "prompt_text_column": null}