© 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. 28626
    Prev
    Next
    INDEX
    Explanations

    numeric values and measurements

    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
    186
    +0.18
    1.0%
    307
    +0.15
    0.9%
    237
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    307
    +0.18
    0.04
    82
    +0.15
    0.04
    136
    +0.12
    -0.00
    Negative Logits
     compelling
    -1.55
    ço
    -1.35
     messenger
    -1.34
    áĢ¸
    -1.33
     authority
    -1.29
    oj
    -1.27
    City
    -1.27
     mandate
    -1.25
    ”:
    -1.25
     vm
    -1.24
    POSITIVE LOGITS
    Īĺ
    3.52
    ĥ½
    2.69
    «
    2.60
    ª
    2.48
    ĥ
    2.46
    ↵
    2.44
    č↵č↵
    2.44
    2.44
    ↵↵
    2.44
    2.44
    Activations Density 0.545%

    No Known Activations