© 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. Residuals
    5. 2-RES-SM
    6. 12272
    Prev
    Next
    INDEX
    Explanations

    technical instructions and experiences related to using devices and software

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/2-res-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_resid_post
    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
    333
    +0.12
    0.7%
    436
    +0.11
    0.6%
    31
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    56
    +0.12
    -0.02
    287
    +0.11
    0.13
    365
    +0.11
    0.07
    Negative Logits
    ¯
    -2.61
    «
    -2.57
    ĥ
    -2.48
    ¶
    -2.41
    ¾
    -2.40
    ¬
    -2.40
    »¿
    -2.39
    Ĥ¬
    -2.35
    ı
    -2.25
    İ
    -2.22
    POSITIVE LOGITS
     tutorials
    1.74
    "}](#
    1.72
     drawer
    1.69
    heets
    1.66
     lately
    1.66
    idenote
    1.59
     docs
    1.56
    ULTS
    1.52
     yesterday
    1.50
     myself
    1.49
    Activations Density 2.958%

    No Known Activations