Update tiny models for GlmAsrForConditionalGeneration

chat_template.jinja

@@ -0,0 +1,32 @@
+{%- macro to_text(content) -%}
+{%- if content is string -%}
+{{- content -}}
+{%- elif content is iterable and content is not mapping -%}
+{%- for item in content -%}
+{%- if item is mapping and item.type == 'text' and item.text is defined -%}
+{{- item.text -}}
+{%- elif item is mapping and (item.type == 'audio' or 'audio' in item) -%}
+<|begin_of_audio|><|pad|><|end_of_audio|><|user|>
+{% elif item is string -%}
+{{- item -}}
+{%- endif -%}
+{%- endfor -%}
+{%- else -%}
+{{- content -}}
+{%- endif -%}
+{%- endmacro -%}
+{%- for m in messages -%}
+{%- if m.role == 'system' -%}
+<|system|>
+{{ to_text(m.content) | trim }}
+{%- elif m.role == 'user' -%}
+<|user|>
+{{ to_text(m.content) | trim }}
+{%- elif m.role == 'assistant' -%}
+<|assistant|>
+{{ to_text(m.content) | trim }}
+{%- endif -%}
+{%- endfor -%}
+{%- if add_generation_prompt -%}
+<|assistant|>
+{% endif -%}

config.json

@@ -0,0 +1,58 @@
+{
+  "architectures": [
+    "GlmAsrForConditionalGeneration"
+  ],
+  "audio_config": {
+    "attention_dropout": 0.0,
+    "hidden_act": "gelu",
+    "hidden_size": 32,
+    "initializer_range": 0.02,
+    "intermediate_size": 32,
+    "max_position_embeddings": 512,
+    "model_type": "glmasr_encoder",
+    "num_attention_heads": 2,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "num_mel_bins": 80,
+    "partial_rotary_factor": 0.5,
+    "rope_parameters": {
+      "partial_rotary_factor": 0.5,
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    }
+  },
+  "audio_token_id": 0,
+  "dtype": "float32",
+  "model_type": "glmasr",
+  "pad_token_id": 59246,
+  "projector_hidden_act": "gelu",
+  "text_config": {
+    "attention_bias": false,
+    "attention_dropout": 0.0,
+    "bos_token_id": 1,
+    "eos_token_id": 59246,
+    "head_dim": 8,
+    "hidden_act": "gelu",
+    "hidden_size": 32,
+    "initializer_range": 0.02,
+    "intermediate_size": 32,
+    "max_position_embeddings": 512,
+    "mlp_bias": false,
+    "model_type": "llama",
+    "num_attention_heads": 2,
+    "num_hidden_layers": 2,
+    "num_key_value_heads": 2,
+    "pad_token_id": 59246,
+    "pretraining_tp": 1,
+    "rms_norm_eps": 1e-06,
+    "rope_parameters": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    },
+    "tie_word_embeddings": false,
+    "use_cache": true,
+    "vocab_size": 59263
+  },
+  "tie_word_embeddings": true,
+  "transformers_version": "5.16.0.dev0"
+}

generation_config.json

@@ -0,0 +1,10 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 59246,
+  "output_attentions": false,
+  "output_hidden_states": false,
+  "pad_token_id": 59246,
+  "transformers_version": "5.16.0.dev0",
+  "use_cache": true
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80328d9f9ff6b4033095caa8fb3fbff0bf79fa943c3c25413e2060330cba673c
+size 7745560

preprocessor_config.json

@@ -0,0 +1,14 @@
+{
+  "chunk_length": 30,
+  "dither": 0.0,
+  "feature_extractor_type": "WhisperFeatureExtractor",
+  "feature_size": 128,
+  "hop_length": 160,
+  "n_fft": 400,
+  "n_samples": 480000,
+  "nb_max_frames": 3000,
+  "padding_side": "right",
+  "padding_value": 0.0,
+  "return_attention_mask": false,
+  "sampling_rate": 16000
+}

tokenizer_config.json

@@ -0,0 +1,38 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": false,
+  "do_lower_case": false,
+  "eos_token": "<|endoftext|>",
+  "extra_special_tokens": [
+    "<|endoftext|>",
+    "[MASK]",
+    "[gMASK]",
+    "[sMASK]",
+    "<sop>",
+    "<eop>",
+    "<|system|>",
+    "<|user|>",
+    "<|assistant|>",
+    "<|observation|>",
+    "<|begin_of_image|>",
+    "<|end_of_image|>",
+    "<|begin_of_video|>",
+    "<|end_of_video|>",
+    "<|pad|>",
+    "<|begin_of_audio|>",
+    "<|end_of_audio|>"
+  ],
+  "is_local": true,
+  "local_files_only": false,
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 65536,
+  "model_specific_special_tokens": {},
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "GlmAsrProcessor",
+  "remove_space": false,
+  "tokenizer_class": "TokenizersBackend"
+}