© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. MLP Post
    9. 2-MLP-SM
    11. 26673
    Prev
    Next
    INDEX
    Explanations

    code and function-related expressions indicating file and data handling processes

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.2.hook_mlp_out
    Hook Layer
    2
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    178
    +0.15
    0.8%
    30
    +0.11
    0.6%
    282
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.15
    0.12
    30
    +0.11
    0.04
    178
    +0.10
    0.09
    Negative Logits
     manually
    -1.95
     Stats
    -1.64
     synaptic
    -1.58
     administr
    -1.56
     logar
    -1.51
     login
    -1.50
     upd
    -1.49
     integr
    -1.46
     vet
    -1.44
    uncher
    -1.42
    POSITIVE LOGITS
    Ĩ
    2.26
    ½
    2.08
    »¿
    1.98
    ĥ½
    1.86
    ¼
    1.81
    Ĥ
    1.79
    Ħ
    1.79
    ´
    1.74
    »
    1.61
    ĺ
    1.55
    Activations Density 3.104%

    No Known Activations