© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 28628
    Prev
    Next
    INDEX
    Explanations

    instances of content related to creation or production

    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
    150
    +0.13
    0.7%
    376
    +0.12
    0.7%
    15
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.13
    0.02
    15
    +0.12
    0.09
    277
    +0.12
    0.07
    Negative Logits
    »¿
    -3.44
    į
    -2.94
    ĭ
    -2.85
    ĥ
    -2.84
    Ĩ
    -2.81
    ij
    -2.73
    Ģ
    -2.71
    ¢
    -2.59
    Ļª
    -2.56
    Ĥ¬
    -2.47
    POSITIVE LOGITS
    leine
    1.90
     by
    1.75
    using
    1.63
     annealing
    1.58
    irectory
    1.57
    InstanceState
    1.56
     versions
    1.54
    ications
    1.52
    uate
    1.51
    ocument
    1.46
    Activations Density 1.001%

    No Known Activations