© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. MLP Post
    5. 2-MLP-SM
    6. 25365
    Prev
    Next
    INDEX
    Explanations

    block structure and control flow statements in programming code

    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
    313
    +0.16
    0.9%
    429
    +0.13
    0.7%
    378
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    313
    +0.16
    0.05
    378
    +0.13
    0.04
    385
    +0.13
    0.05
    Negative Logits
     ensued
    -1.62
     upset
    -1.46
    ?).
    -1.41
     sympath
    -1.41
     Gulf
    -1.39
     RH
    -1.36
    rac
    -1.31
     detriment
    -1.31
     ranc
    -1.30
     interest
    -1.28
    POSITIVE LOGITS
    :]
    1.59
    osi
    1.55
    itals
    1.46
    AtIndex
    1.41
    anges
    1.41
    [\#
    1.38
    zes
    1.37
    ül
    1.36
    Ļª
    1.34
     copies
    1.33
    Activations Density 0.203%

    No Known Activations