optimum-internal-testing-user commited on
Commit
72abfd2
·
verified ·
1 Parent(s): f86fb73

Upload folder using huggingface_hub

Browse files
Files changed (3) hide show
  1. config.json +37 -0
  2. model.pt +3 -0
  3. neuron_config.json +29 -0
config.json ADDED
@@ -0,0 +1,37 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "LlamaForCausalLM"
4
+ ],
5
+ "attention_bias": false,
6
+ "attention_dropout": 0.0,
7
+ "bos_token_id": 128000,
8
+ "eos_token_id": 128009,
9
+ "head_dim": 64,
10
+ "hidden_act": "silu",
11
+ "hidden_size": 2048,
12
+ "initializer_range": 0.02,
13
+ "intermediate_size": 8192,
14
+ "max_position_embeddings": 131072,
15
+ "mlp_bias": false,
16
+ "model_type": "llama",
17
+ "num_attention_heads": 32,
18
+ "num_hidden_layers": 16,
19
+ "num_key_value_heads": 8,
20
+ "pad_token_id": 128004,
21
+ "pretraining_tp": 1,
22
+ "rms_norm_eps": 1e-05,
23
+ "rope_scaling": {
24
+ "factor": 32.0,
25
+ "high_freq_factor": 4.0,
26
+ "low_freq_factor": 1.0,
27
+ "original_max_position_embeddings": 8192,
28
+ "rope_type": "llama3"
29
+ },
30
+ "rope_theta": 500000.0,
31
+ "tie_word_embeddings": true,
32
+ "torch_dtype": "bfloat16",
33
+ "transformers_version": "4.55.4",
34
+ "unsloth_fixed": true,
35
+ "use_cache": true,
36
+ "vocab_size": 128256
37
+ }
model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:25847e99908365bc65feec6cf4a3ddfce26ad863882c0288d098bf08fb4ba165
3
+ size 34560801
neuron_config.json ADDED
@@ -0,0 +1,29 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_serialized_key": "NxDNeuronConfig",
3
+ "batch_size": 1,
4
+ "capacity_factor": null,
5
+ "checkpoint_id": "unsloth/Llama-3.2-1B-Instruct",
6
+ "checkpoint_revision": null,
7
+ "continuous_batching": false,
8
+ "enable_bucketing": false,
9
+ "ep_degree": 1,
10
+ "fused_qkv": false,
11
+ "glu_mlp": true,
12
+ "local_ranks_size": 2,
13
+ "logical_nc_config": 1,
14
+ "max_batch_size": 1,
15
+ "max_context_length": 4096,
16
+ "max_topk": 256,
17
+ "n_active_tokens": 4096,
18
+ "neuronxcc_version": "2.19.8089.0+8ab9f450",
19
+ "on_device_sampling": false,
20
+ "optimum_neuron_version": "0.3.1.dev5",
21
+ "output_logits": false,
22
+ "pp_degree": 1,
23
+ "sequence_length": 4096,
24
+ "speculation_length": 0,
25
+ "start_rank_id": 0,
26
+ "target": null,
27
+ "torch_dtype": "bfloat16",
28
+ "tp_degree": 2
29
+ }