treyleo16/ brainrot:67

15 4 weeks ago

A model that loves: SIiIiX SeVeN.

135m
... /
model
2215528cc8b9 · 534kB
    Metadata
  • general.architecture
    llama
  • general.file_type
    Q4_0
  • llama.attention.head_count
    4
  • llama.attention.head_count_kv
    4
  • llama.attention.layer_norm_rms_epsilon
    1e-05
  • llama.block_count
    4
  • llama.context_length
    512
  • llama.embedding_length
    128
  • llama.feed_forward_length
    352
  • llama.rope.dimension_count
    32
  • llama.rope.freq_base
    10000
  • llama.vocab_size
    260
  • tokenizer.ggml.add_bos_token
    false
  • tokenizer.ggml.add_eos_token
    false
  • tokenizer.ggml.bos_token_id
    256
  • tokenizer.ggml.eos_token_id
    257
  • tokenizer.ggml.eot_token_id
    257
  • tokenizer.ggml.merges
    []
  • tokenizer.ggml.model
    gpt2
  • tokenizer.ggml.padding_token_id
    257
  • tokenizer.ggml.pre
    default
  • tokenizer.ggml.token_type
    [1, 1, 1, 1, 1, ...]
  • tokenizer.ggml.tokens
    [Ā, ā, Ă, ă, Ą, ...]
    • Tensor
  • token_embd.weight
    Q8_0
    [128, 260]
    • blk.0
  • blk.0.attn_k.weight
    Q4_0
    [128, 128]
  • blk.0.attn_norm.weight
    F32
    [128]
  • blk.0.attn_output.weight
    Q4_0
    [128, 128]
  • blk.0.attn_q.weight
    Q4_0
    [128, 128]
  • blk.0.attn_v.weight
    Q4_0
    [128, 128]
  • blk.0.ffn_down.weight
    Q4_0
    [352, 128]
  • blk.0.ffn_gate.weight
    Q4_0
    [128, 352]
  • blk.0.ffn_norm.weight
    F32
    [128]
  • blk.0.ffn_up.weight
    Q4_0
    [128, 352]
    • blk.1
  • blk.1.attn_k.weight
    Q4_0
    [128, 128]
  • blk.1.attn_norm.weight
    F32
    [128]
  • blk.1.attn_output.weight
    Q4_0
    [128, 128]
  • blk.1.attn_q.weight
    Q4_0
    [128, 128]
  • blk.1.attn_v.weight
    Q4_0
    [128, 128]
  • blk.1.ffn_down.weight
    Q4_0
    [352, 128]
  • blk.1.ffn_gate.weight
    Q4_0
    [128, 352]
  • blk.1.ffn_norm.weight
    F32
    [128]
  • blk.1.ffn_up.weight
    Q4_0
    [128, 352]
    • blk.2
  • blk.2.attn_k.weight
    Q4_0
    [128, 128]
  • blk.2.attn_norm.weight
    F32
    [128]
  • blk.2.attn_output.weight
    Q4_0
    [128, 128]
  • blk.2.attn_q.weight
    Q4_0
    [128, 128]
  • blk.2.attn_v.weight
    Q4_0
    [128, 128]
  • blk.2.ffn_down.weight
    Q4_0
    [352, 128]
  • blk.2.ffn_gate.weight
    Q4_0
    [128, 352]
  • blk.2.ffn_norm.weight
    F32
    [128]
  • blk.2.ffn_up.weight
    Q4_0
    [128, 352]
    • blk.3
  • blk.3.attn_k.weight
    Q4_0
    [128, 128]
  • blk.3.attn_norm.weight
    F32
    [128]
  • blk.3.attn_output.weight
    Q4_0
    [128, 128]
  • blk.3.attn_q.weight
    Q4_0
    [128, 128]
  • blk.3.attn_v.weight
    Q4_0
    [128, 128]
  • blk.3.ffn_down.weight
    Q4_0
    [352, 128]
  • blk.3.ffn_gate.weight
    Q4_0
    [128, 352]
  • blk.3.ffn_norm.weight
    F32
    [128]
  • blk.3.ffn_up.weight
    Q4_0
    [128, 352]
  • output.weight
    Q8_0
    [128, 260]
  • output_norm.weight
    F32
    [128]