Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 28830
    Prev
    Next
    INDEX
    Explanations

    phrases related to delivering or providing services

    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
    148
    +0.16
    0.9%
    246
    +0.12
    0.7%
    351
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    148
    +0.16
    0.01
    246
    +0.12
    0.02
    490
    +0.12
    0.02
    Negative Logits
    isme
    -1.50
    seen
    -1.41
    COP
    -1.41
     Pradesh
    -1.39
    iska
    -1.39
    "}](#
    -1.35
    ism
    -1.34
    omorphism
    -1.31
    ifndef
    -1.31
    faced
    -1.31
    POSITIVE LOGITS
    ents
    1.86
    ables
    1.77
    antic
    1.66
    ancers
    1.53
    ance
    1.51
    imental
    1.50
    acity
    1.50
     delivery
    1.48
    able
    1.46
    ency
    1.46
    Activations Density 0.125%

    No Known Activations