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

    Neuronpedia

    APICircuit 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. 11538
    Prev
    Next
    INDEX
    Explanations

    terms related to file formats and their specifications

    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
    100
    +0.17
    0.9%
    217
    +0.16
    0.9%
    385
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    215
    +0.17
    0.03
    100
    +0.16
    0.02
    385
    +0.13
    0.01
    Negative Logits
    ¼
    -2.00
    ĻĤ
    -1.83
    ±
    -1.81
    ½
    -1.79
    dule
    -1.79
    ¸
    -1.76
    §
    -1.72
    ¤
    -1.72
    ¹
    -1.72
    sell
    -1.71
    POSITIVE LOGITS
    ting
    2.21
    ters
    2.08
    imetry
    2.04
    inally
    2.03
    imetric
    1.92
    eline
    1.88
    table
    1.78
    aneous
    1.77
    ional
    1.72
    etting
    1.69
    Activations Density 0.138%

    No Known Activations