Vocabulary projection
×learnedself.lm_head(hidden_states)
Project each normalized feature vector into vocabulary logits.
- in
- [B batch, T positions, 7168 d]
- out
- [B batch, T positions, 163840 vocabulary]
- parameters
- 1,174,405,120
Logits are cast to float32 (modeling_deepseek.py 1615–1616).
Parameters
1,174,405,120 parameters per instance.
| Component | Calculation | Parameters |
|---|---|---|
| Weight matrix | 1,174,405,120 | |
| Per-instance total | 1,174,405,120 | |
The reference module holds the weight output by input, [out, in].
Derived from the configured matrix dimensions; the model total matches the checkpoint’s tensors.
Explanation
Each entry scores one vocabulary token as the next token for its position. The softmax and the choice of token happen outside the model.
Every position’s logits are computed; generation reads the last.
Codemodeling_deepseek.py 1360–1616 · config.json 1–68
self.embed_tokens = nn.Embedding(
config.vocab_size, config.hidden_size, self.padding_idx
)
self.layers = nn.ModuleList(
[
DeepseekV3DecoderLayer(config, layer_idx)
for layer_idx in range(config.num_hidden_layers)
]
)
self._use_flash_attention_2 = config._attn_implementation == "flash_attention_2"
self.norm = DeepseekV3RMSNorm(config.hidden_size, eps=config.rms_norm_eps) past_key_values_length = 0
if use_cache:
use_legacy_cache = not isinstance(past_key_values, Cache)
if use_legacy_cache:
past_key_values = DynamicCache.from_legacy_cache(past_key_values)
past_key_values_length = past_key_values.get_usable_length(seq_length)
if position_ids is None:
device = input_ids.device if input_ids is not None else inputs_embeds.device
position_ids = torch.arange(
past_key_values_length,
seq_length + past_key_values_length,
dtype=torch.long,
device=device,
)
position_ids = position_ids.unsqueeze(0) if self._use_flash_attention_2:
# 2d mask is passed through the layers
attention_mask = (
attention_mask
if (attention_mask is not None and 0 in attention_mask)
else None
)
else:
# 4d mask is passed through the layers
attention_mask = _prepare_4d_causal_attention_mask(
attention_mask,
(batch_size, seq_length),
inputs_embeds,
past_key_values_length,
) self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False) hidden_states = outputs[0]
logits = self.lm_head(hidden_states)
logits = logits.float(){
"architectures": [
"DeepseekV3ForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "configuration_deepseek.DeepseekV3Config",
"AutoModel": "modeling_deepseek.DeepseekV3Model",
"AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
},
"aux_loss_alpha": 0.001,
"bos_token_id": 163584,
"eos_token_id": 163585,
"first_k_dense_replace": 1,
"hidden_act": "silu",
"hidden_size": 7168,
"initializer_range": 0.02,
"intermediate_size": 18432,
"kv_lora_rank": 512,
"max_position_embeddings": 131072,
"model_type": "kimi_k2",
"moe_intermediate_size": 2048,
"moe_layer_freq": 1,
"n_group": 1,
"n_routed_experts": 384,
"n_shared_experts": 1,
"norm_topk_prob": true,
"num_attention_heads": 64,
"num_experts_per_tok": 8,
"num_hidden_layers": 61,
"num_key_value_heads": 64,
"num_nextn_predict_layers": 0,
"pretraining_tp": 1,
"q_lora_rank": 1536,
"qk_nope_head_dim": 128,
"qk_rope_head_dim": 64,
"quantization_config": {
"activation_scheme": "dynamic",
"fmt": "e4m3",
"quant_method": "fp8",
"weight_block_size": [
128,
128
]
},
"rms_norm_eps": 1e-06,
"rope_theta": 50000.0,
"routed_scaling_factor": 2.827,
"rope_scaling": {
"beta_fast": 1.0,
"beta_slow": 1.0,
"factor": 32.0,
"mscale": 1.0,
"mscale_all_dim": 1.0,
"original_max_position_embeddings": 4096,
"type": "yarn"
},
"scoring_func": "sigmoid",
"seq_aux": true,
"tie_word_embeddings": false,
"topk_group": 1,
"topk_method": "noaux_tc",
"torch_dtype": "bfloat16",
"transformers_version": "4.48.3",
"use_cache": true,
"v_head_dim": 128,
"vocab_size": 163840Sources2
Direct evidence
The model embeds tokens, numbers positions from the cached length, builds a 4D causal mask, applies all blocks, normalizes and projects to vocabulary logits with an untied head, computing every position’s logits in float32.
transformer · moonshotai/Kimi-K2-Instruct / modeling_deepseek.py
Pinned revision fd1984e2b7a3 ↗ · lines 1360–1616
Read code excerpt
self.embed_tokens = nn.Embedding(
config.vocab_size, config.hidden_size, self.padding_idx
)
self.layers = nn.ModuleList(
[
DeepseekV3DecoderLayer(config, layer_idx)
for layer_idx in range(config.num_hidden_layers)
]
)
self._use_flash_attention_2 = config._attn_implementation == "flash_attention_2"
self.norm = DeepseekV3RMSNorm(config.hidden_size, eps=config.rms_norm_eps)… lines 1371–1422 omitted …
past_key_values_length = 0
if use_cache:
use_legacy_cache = not isinstance(past_key_values, Cache)
if use_legacy_cache:
past_key_values = DynamicCache.from_legacy_cache(past_key_values)
past_key_values_length = past_key_values.get_usable_length(seq_length)
if position_ids is None:
device = input_ids.device if input_ids is not None else inputs_embeds.device
position_ids = torch.arange(
past_key_values_length,
seq_length + past_key_values_length,
dtype=torch.long,
device=device,
)
position_ids = position_ids.unsqueeze(0)… lines 1439–1442 omitted …
if self._use_flash_attention_2:
# 2d mask is passed through the layers
attention_mask = (
attention_mask
if (attention_mask is not None and 0 in attention_mask)
else None
)
else:
# 4d mask is passed through the layers
attention_mask = _prepare_4d_causal_attention_mask(
attention_mask,
(batch_size, seq_length),
inputs_embeds,
past_key_values_length,
)… lines 1458–1521 omitted …
self.lm_head = nn.Linear(config.hidden_size, config.vocab_size, bias=False)… lines 1523–1613 omitted …
hidden_states = outputs[0]
logits = self.lm_head(hidden_states)
logits = logits.float()Direct evidence
The released configuration: width, layers, dense layers, heads, routed and shared experts, routing (sigmoid scores, one expert group, top-8 of 384, route scale 2.827), MLA ranks and head widths, vocabulary, YaRN RoPE.
config · moonshotai/Kimi-K2-Instruct / config.json
Pinned revision fd1984e2b7a3 ↗ · lines 1–68
Read code excerpt
{
"architectures": [
"DeepseekV3ForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "configuration_deepseek.DeepseekV3Config",
"AutoModel": "modeling_deepseek.DeepseekV3Model",
"AutoModelForCausalLM": "modeling_deepseek.DeepseekV3ForCausalLM"
},
"aux_loss_alpha": 0.001,
"bos_token_id": 163584,
"eos_token_id": 163585,
"first_k_dense_replace": 1,
"hidden_act": "silu",
"hidden_size": 7168,
"initializer_range": 0.02,
"intermediate_size": 18432,
"kv_lora_rank": 512,
"max_position_embeddings": 131072,
"model_type": "kimi_k2",
"moe_intermediate_size": 2048,
"moe_layer_freq": 1,
"n_group": 1,
"n_routed_experts": 384,
"n_shared_experts": 1,
"norm_topk_prob": true,
"num_attention_heads": 64,
"num_experts_per_tok": 8,
"num_hidden_layers": 61,
"num_key_value_heads": 64,
"num_nextn_predict_layers": 0,
"pretraining_tp": 1,
"q_lora_rank": 1536,
"qk_nope_head_dim": 128,
"qk_rope_head_dim": 64,
"quantization_config": {
"activation_scheme": "dynamic",
"fmt": "e4m3",
"quant_method": "fp8",
"weight_block_size": [
128,
128
]
},
"rms_norm_eps": 1e-06,
"rope_theta": 50000.0,
"routed_scaling_factor": 2.827,
"rope_scaling": {
"beta_fast": 1.0,
"beta_slow": 1.0,
"factor": 32.0,
"mscale": 1.0,
"mscale_all_dim": 1.0,
"original_max_position_embeddings": 4096,
"type": "yarn"
},
"scoring_func": "sigmoid",
"seq_aux": true,
"tie_word_embeddings": false,
"topk_group": 1,
"topk_method": "noaux_tc",
"torch_dtype": "bfloat16",
"transformers_version": "4.48.3",
"use_cache": true,
"v_head_dim": 128,
"vocab_size": 163840