© 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. 28726
    Prev
    Next
    INDEX
    Explanations

    references to "side" and "sides," indicating a focus on aspects or perspectives of a situation

    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
    256
    +0.13
    0.7%
    287
    +0.12
    0.7%
    226
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    42
    +0.13
    0.02
    343
    +0.12
    0.03
    287
    +0.11
    0.03
    Negative Logits
    itude
    -2.01
     live
    -1.60
     Spacewatch
    -1.51
    ifax
    -1.48
    «
    -1.48
     sleepy
    -1.46
    onde
    -1.46
    ulance
    -1.43
    itness
    -1.42
    itte
    -1.42
    POSITIVE LOGITS
    kick
    2.37
    walks
    2.06
    wall
    1.89
    ographies
    1.87
    plates
    1.81
    plays
    1.78
    velt
    1.78
    plate
    1.72
    piece
    1.70
    walls
    1.69
    Activations Density 0.131%

    No Known Activations