Upload folder using huggingface_hub

chat_template.jinja

@@ -0,0 +1,33 @@
+{% if not messages[0]['role'] == 'system' %}### INSTRUCTION:
+You are a precise Content Moderation Classifier. Your mission is to evaluate text for toxicity, bias, and hate speech. You must distinguish between overt abuse and "implicit hate"—text that may seem linguistically neutral but carries harmful, stereotypical, or dehumanizing intent toward protected groups.
+
+## NOT ALLOWED
+You must flag and categorize the following as Violations:
+
+Identity-Based Hate: Any content targeting individuals or groups based on: Women, LGBTQ+, Mental/Physical Disorders, Race (Black, Chinese, Asian, Native American, Latino, Mexican), Religion (Muslim, Jewish), or Middle Eastern origin.
+
+Implicit Toxicity: Coded language, "polite" stereotypes, or synthetic generations that imply inferiority or promote exclusion without using explicit slurs.
+
+Dehumanization: Language comparing protected groups to animals, insects, diseases, or inanimate objects.
+
+Aggressive Toxicity: Direct insults, derogatory terms, harmful slang, and threatening language.
+
+Bias Reinforcement: Statements that generalize negative traits or "harmful tropes" to an entire demographic.
+
+## ALLOWED
+You must classify the following as Benign:
+
+Neutral Reference: Factual or descriptive mentions of protected groups (e.g., "The history of Latino art").
+
+Educational/Clinical Context: Objective discussion of disabilities, mental health, or religious practices without judgmental or biased framing.
+
+Self-Identification: Individuals using identity terms to describe themselves or their own community experiences in a non-harmful manner.
+
+Counter-Speech: Content that explicitly challenges or denounces hate speech, provided it does not repeat the toxicity in a gratuitous way.
+{% endif %}{% for message in messages %}{% if message['role'] == 'system' %}### INSTRUCTION:
+{{ message['content'] }}
+{% elif message['role'] == 'user' %}### INPUT:
+{{ message['content'] }}
+{% elif message['role'] == 'assistant' %}### OUTPUT:
+{{ message['content'] }}
+{% endif %}{% endfor %}

config.json

@@ -0,0 +1,41 @@
+{
+  "activation_function": "gelu_new",
+  "add_cross_attention": false,
+  "architectures": [
+    "GPT2LMHeadModel"
+  ],
+  "attn_pdrop": 0.1,
+  "bos_token_id": 50256,
+  "dtype": "float32",
+  "embd_pdrop": 0.1,
+  "eos_token_id": 50256,
+  "initializer_range": 0.02,
+  "layer_norm_epsilon": 1e-05,
+  "model_type": "gpt2",
+  "n_ctx": 1024,
+  "n_embd": 768,
+  "n_head": 12,
+  "n_inner": null,
+  "n_layer": 12,
+  "n_positions": 1024,
+  "pad_token_id": 50256,
+  "reorder_and_upcast_attn": false,
+  "resid_pdrop": 0.1,
+  "scale_attn_by_inverse_layer_idx": false,
+  "scale_attn_weights": true,
+  "summary_activation": null,
+  "summary_first_dropout": 0.1,
+  "summary_proj_to_labels": true,
+  "summary_type": "cls_index",
+  "summary_use_proj": true,
+  "task_specific_params": {
+    "text-generation": {
+      "do_sample": true,
+      "max_length": 50
+    }
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.2.0",
+  "use_cache": true,
+  "vocab_size": 50257
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 50256,
+  "eos_token_id": 50256,
+  "transformers_version": "5.2.0"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8418ae9424cdab8a38dbdd3bc25acad8e484e4e96fc1f4076f2d1fe8cc97037a
+size 497777468

tokenizer_config.json

@@ -0,0 +1,12 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": "<|endoftext|>",
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "is_local": false,
+  "model_max_length": 1024,
+  "pad_token": "<|endoftext|>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "<|endoftext|>"
+}