© 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. 25560
    Prev
    Next
    INDEX
    Explanations

    specific programming or technical language constructs and keywords

    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
    150
    +0.14
    0.8%
    271
    +0.12
    0.6%
    417
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    279
    +0.14
    0.08
    462
    +0.12
    0.01
    445
    +0.11
    0.00
    Negative Logits
    *](#
    -1.86
     Redistributions
    -1.72
    ción
    -1.67
    orf
    -1.64
    olation
    -1.62
    WORD
    -1.60
    iter
    -1.58
    orescence
    -1.55
    ^](#
    -1.51
    ophagy
    -1.49
    POSITIVE LOGITS
     flagship
    1.82
    ĥ
    1.80
     anchors
    1.77
    ª
    1.76
    ħ
    1.71
     affiliates
    1.66
    ŀ
    1.66
     antagonists
    1.56
     ambassador
    1.56
    ®
    1.55
    Activations Density 2.612%

    No Known Activations