© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    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. 1-RES-SM
    11. 27594
    Prev
    Next
    INDEX
    Explanations

    numerical values in various contexts

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    6
    +0.10
    0.5%
    470
    +0.09
    0.5%
    286
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    203
    +0.10
    0.07
    136
    +0.09
    0.00
    246
    +0.09
    0.05
    Negative Logits
    atics
    -1.53
    ements
    -1.44
    ments
    -1.44
     ("
    -1.42
    .....
    -1.37
    ,
    -1.36
    tees
    -1.35
    ceral
    -1.34
    bestos
    -1.33
     };
    -1.32
    POSITIVE LOGITS
    ]{}\_
    1.65
    &=&\
    1.57
    ]{}\
    1.57
    ]{}\_[
    1.54
    ]{}
    1.52
    ``
    1.50
    hin
    1.50
    ®
    1.49
    {{{\
    1.48
    mapsto
    1.47
    Activations Density 0.483%

    No Known Activations