Upload model
Browse files- config.json +31 -3
- generation_config.json +5 -1
- model.safetensors +2 -2
config.json
CHANGED
|
@@ -1,19 +1,47 @@
|
|
| 1 |
{
|
|
|
|
| 2 |
"architectures": [
|
| 3 |
"MultiTaskModel"
|
| 4 |
],
|
|
|
|
|
|
|
| 5 |
"base_model_name": "HuggingFaceTB/SmolLM2-135M-Instruct",
|
|
|
|
| 6 |
"debug": true,
|
|
|
|
|
|
|
|
|
|
| 7 |
"hidden_size": 576,
|
| 8 |
-
"
|
| 9 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
"pad_token_id": 2,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 11 |
"sequence_lengths": {
|
| 12 |
"formula": 128,
|
| 13 |
"prompt": 256,
|
| 14 |
"rationale": 384
|
| 15 |
},
|
| 16 |
"torch_dtype": "float32",
|
| 17 |
-
"
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 18 |
"vocab_size": 49152
|
| 19 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "HuggingFaceTB/SmolLM2-135M-Instruct",
|
| 3 |
"architectures": [
|
| 4 |
"MultiTaskModel"
|
| 5 |
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
"base_model_name": "HuggingFaceTB/SmolLM2-135M-Instruct",
|
| 9 |
+
"bos_token_id": 1,
|
| 10 |
"debug": true,
|
| 11 |
+
"eos_token_id": 2,
|
| 12 |
+
"head_dim": 64,
|
| 13 |
+
"hidden_act": "silu",
|
| 14 |
"hidden_size": 576,
|
| 15 |
+
"initializer_range": 0.041666666666666664,
|
| 16 |
+
"intermediate_size": 1536,
|
| 17 |
+
"is_llama_config": true,
|
| 18 |
+
"max_position_embeddings": 8192,
|
| 19 |
+
"mlp_bias": false,
|
| 20 |
+
"model_type": "llama",
|
| 21 |
+
"num_attention_heads": 9,
|
| 22 |
+
"num_choices": 5,
|
| 23 |
+
"num_hidden_layers": 30,
|
| 24 |
+
"num_key_value_heads": 3,
|
| 25 |
+
"output_hidden_states": true,
|
| 26 |
"pad_token_id": 2,
|
| 27 |
+
"pretraining_tp": 1,
|
| 28 |
+
"rms_norm_eps": 1e-05,
|
| 29 |
+
"rope_interleaved": false,
|
| 30 |
+
"rope_scaling": null,
|
| 31 |
+
"rope_theta": 100000,
|
| 32 |
"sequence_lengths": {
|
| 33 |
"formula": 128,
|
| 34 |
"prompt": 256,
|
| 35 |
"rationale": 384
|
| 36 |
},
|
| 37 |
"torch_dtype": "float32",
|
| 38 |
+
"transformers.js_config": {
|
| 39 |
+
"kv_cache_dtype": {
|
| 40 |
+
"fp16": "float16",
|
| 41 |
+
"q4f16": "float16"
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"transformers_version": "4.47.1",
|
| 45 |
+
"use_cache": true,
|
| 46 |
"vocab_size": 49152
|
| 47 |
}
|
generation_config.json
CHANGED
|
@@ -1,5 +1,9 @@
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
|
|
|
|
|
|
|
|
|
| 3 |
"pad_token_id": 2,
|
| 4 |
-
"
|
|
|
|
| 5 |
}
|
|
|
|
| 1 |
{
|
| 2 |
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 1,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"output_hidden_states": true,
|
| 6 |
"pad_token_id": 2,
|
| 7 |
+
"return_dict_in_generate": true,
|
| 8 |
+
"transformers_version": "4.47.1"
|
| 9 |
}
|
model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:96accbdb7ea5ca8dfead1e819a248a33f606dd0b981071874d6cc4314241bb32
|
| 3 |
+
size 1784207004
|