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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 11951
    Prev
    Next
    INDEX
    Explanations

    geographical locations and their descriptions

    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
    26
    +0.12
    0.7%
    9
    +0.11
    0.6%
    156
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    248
    +0.12
    0.10
    98
    +0.11
    -0.03
    26
    +0.11
    0.03
    Negative Logits
     analys
    -2.18
     fulfil
    -2.17
     tumour
    -2.15
     realise
    -2.05
     flavour
    -1.99
     anaest
    -1.97
     programme
    -1.96
     obj
    -1.95
     analyse
    -1.94
    isation
    -1.92
    POSITIVE LOGITS
     statewide
    2.73
     County
    2.41
     zoning
    2.33
     Highway
    2.31
     downtown
    2.31
     Township
    2.30
     Southwest
    2.28
     Interstate
    2.27
    County
    2.26
     homeowners
    2.25
    Activations Density 3.165%

    No Known Activations