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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 26049
    Prev
    Next
    INDEX
    Explanations

    references to studies and research analyses

    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
    388
    +0.15
    0.8%
    459
    +0.14
    0.8%
    392
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    388
    +0.15
    0.02
    459
    +0.14
    0.03
    219
    +0.12
    0.02
    Negative Logits
    ³
    -2.23
    Ļ
    -2.21
    Ń
    -2.18
    ĻĤ
    -2.11
    ¡
    -2.07
    µ
    -2.02
    »
    -1.96
    ķ
    -1.93
    ª
    -1.91
    ħ
    -1.88
    POSITIVE LOGITS
    procedure
    1.63
    heer
    1.61
    lio
    1.49
    "}](#
    1.48
     âĢķ
    1.48
     invention
    1.40
     partnership
    1.36
    rendum
    1.36
    hin
    1.35
    ž
    1.35
    Activations Density 0.089%

    No Known Activations