© 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. Residuals
    9. 2-RES-SM
    11. 28537
    Prev
    Next
    INDEX
    Explanations

    terms associated with recovery and rehabilitation processes

    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
    156
    +0.18
    1.0%
    256
    +0.13
    0.7%
    365
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    239
    +0.18
    0.04
    256
    +0.13
    0.04
    115
    +0.13
    0.02
    Negative Logits
    uke
    -1.82
    iasis
    -1.66
    omorphisms
    -1.64
    omorphism
    -1.59
    chitz
    -1.47
     Race
    -1.46
     dass
    -1.42
    ussi
    -1.39
     typed
    -1.38
    rar
    -1.38
    POSITIVE LOGITS
    InstanceState
    2.00
    giving
    1.71
    ļ
    1.66
    ł
    1.62
    ·¸
    1.56
    stituted
    1.56
    ists
    1.56
    tons
    1.52
     mechanisms
    1.51
    mond
    1.51
    Activations Density 0.299%

    No Known Activations