Text Generation
Transformers
Safetensors
interpgpt
interpretability
mechanistic-interpretability
task-decomposition
small-language-model
transformer-lens
custom_code
Instructions to use connaaa/interpgpt-adhd-23M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use connaaa/interpgpt-adhd-23M with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="connaaa/interpgpt-adhd-23M", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("connaaa/interpgpt-adhd-23M", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use connaaa/interpgpt-adhd-23M with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "connaaa/interpgpt-adhd-23M" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "connaaa/interpgpt-adhd-23M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker
docker model run hf.co/connaaa/interpgpt-adhd-23M
- SGLang
How to use connaaa/interpgpt-adhd-23M with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "connaaa/interpgpt-adhd-23M" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "connaaa/interpgpt-adhd-23M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "connaaa/interpgpt-adhd-23M" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "connaaa/interpgpt-adhd-23M", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }' - Docker Model Runner
How to use connaaa/interpgpt-adhd-23M with Docker Model Runner:
docker model run hf.co/connaaa/interpgpt-adhd-23M
Phase 1 release: InterpGPT matched-pair checkpoint
Browse files- config.json +4 -46
config.json
CHANGED
|
@@ -9,20 +9,13 @@
|
|
| 9 |
"pad_id": 8196,
|
| 10 |
"bias": false,
|
| 11 |
"variant": "adhd",
|
| 12 |
-
"
|
|
|
|
| 13 |
"output_hidden_states": false,
|
| 14 |
-
"
|
| 15 |
"dtype": null,
|
| 16 |
-
"pruned_heads": {},
|
| 17 |
-
"tie_word_embeddings": true,
|
| 18 |
"chunk_size_feed_forward": 0,
|
| 19 |
"is_encoder_decoder": false,
|
| 20 |
-
"is_decoder": false,
|
| 21 |
-
"cross_attention_hidden_size": null,
|
| 22 |
-
"add_cross_attention": false,
|
| 23 |
-
"tie_encoder_decoder": false,
|
| 24 |
-
"architectures": null,
|
| 25 |
-
"finetuning_task": null,
|
| 26 |
"id2label": {
|
| 27 |
"0": "LABEL_0",
|
| 28 |
"1": "LABEL_1"
|
|
@@ -31,48 +24,13 @@
|
|
| 31 |
"LABEL_0": 0,
|
| 32 |
"LABEL_1": 1
|
| 33 |
},
|
| 34 |
-
"task_specific_params": null,
|
| 35 |
"problem_type": null,
|
| 36 |
-
"tokenizer_class": null,
|
| 37 |
-
"prefix": null,
|
| 38 |
-
"bos_token_id": null,
|
| 39 |
-
"pad_token_id": 8196,
|
| 40 |
-
"eos_token_id": null,
|
| 41 |
-
"sep_token_id": null,
|
| 42 |
-
"decoder_start_token_id": null,
|
| 43 |
-
"max_length": 20,
|
| 44 |
-
"min_length": 0,
|
| 45 |
-
"do_sample": false,
|
| 46 |
-
"early_stopping": false,
|
| 47 |
-
"num_beams": 1,
|
| 48 |
-
"temperature": 1.0,
|
| 49 |
-
"top_k": 50,
|
| 50 |
-
"top_p": 1.0,
|
| 51 |
-
"typical_p": 1.0,
|
| 52 |
-
"repetition_penalty": 1.0,
|
| 53 |
-
"length_penalty": 1.0,
|
| 54 |
-
"no_repeat_ngram_size": 0,
|
| 55 |
-
"encoder_no_repeat_ngram_size": 0,
|
| 56 |
-
"bad_words_ids": null,
|
| 57 |
-
"num_return_sequences": 1,
|
| 58 |
-
"output_scores": false,
|
| 59 |
-
"return_dict_in_generate": false,
|
| 60 |
-
"forced_bos_token_id": null,
|
| 61 |
-
"forced_eos_token_id": null,
|
| 62 |
-
"remove_invalid_values": false,
|
| 63 |
-
"exponential_decay_length_penalty": null,
|
| 64 |
-
"suppress_tokens": null,
|
| 65 |
-
"begin_suppress_tokens": null,
|
| 66 |
-
"num_beam_groups": 1,
|
| 67 |
-
"diversity_penalty": 0.0,
|
| 68 |
"_name_or_path": "",
|
| 69 |
-
"
|
| 70 |
"auto_map": {
|
| 71 |
"AutoConfig": "configuration_interpgpt.InterpGPTConfig",
|
| 72 |
"AutoModel": "modeling_interpgpt.InterpGPTModel"
|
| 73 |
},
|
| 74 |
-
"tf_legacy_loss": false,
|
| 75 |
-
"use_bfloat16": false,
|
| 76 |
"model_type": "interpgpt",
|
| 77 |
"output_attentions": false
|
| 78 |
}
|
|
|
|
| 9 |
"pad_id": 8196,
|
| 10 |
"bias": false,
|
| 11 |
"variant": "adhd",
|
| 12 |
+
"transformers_version": "5.5.4",
|
| 13 |
+
"architectures": null,
|
| 14 |
"output_hidden_states": false,
|
| 15 |
+
"return_dict": true,
|
| 16 |
"dtype": null,
|
|
|
|
|
|
|
| 17 |
"chunk_size_feed_forward": 0,
|
| 18 |
"is_encoder_decoder": false,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 19 |
"id2label": {
|
| 20 |
"0": "LABEL_0",
|
| 21 |
"1": "LABEL_1"
|
|
|
|
| 24 |
"LABEL_0": 0,
|
| 25 |
"LABEL_1": 1
|
| 26 |
},
|
|
|
|
| 27 |
"problem_type": null,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
"_name_or_path": "",
|
| 29 |
+
"pad_token_id": 8196,
|
| 30 |
"auto_map": {
|
| 31 |
"AutoConfig": "configuration_interpgpt.InterpGPTConfig",
|
| 32 |
"AutoModel": "modeling_interpgpt.InterpGPTModel"
|
| 33 |
},
|
|
|
|
|
|
|
| 34 |
"model_type": "interpgpt",
|
| 35 |
"output_attentions": false
|
| 36 |
}
|