© 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. Residuals
    9. 2-RES-SM
    11. 12538
    Prev
    Next
    INDEX
    Explanations

    actions and verbs indicating a process or ongoing effort

    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
    323
    +0.21
    1.2%
    442
    +0.12
    0.7%
    161
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    323
    +0.21
    0.10
    243
    +0.12
    0.10
    263
    +0.11
    0.10
    Negative Logits
    į
    -2.00
    ĻĤ
    -1.96
    ĩ
    -1.89
    İ
    -1.77
    ":
    -1.68
    Ļª
    -1.66
    "}](#
    -1.65
    ¤
    -1.64
    Ĵ
    -1.57
    §
    -1.55
    POSITIVE LOGITS
    abouts
    2.06
    coming
    1.62
    ubicin
    1.47
     effect
    1.44
    ialize
    1.41
    stood
    1.39
    agus
    1.38
    holders
    1.38
    wed
    1.34
    bbe
    1.33
    Activations Density 1.408%

    No Known Activations