© 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. MLP Post
    9. 2-MLP-SM
    11. 26088
    Prev
    Next
    INDEX
    Explanations

    identification and references to official legal or organizational entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__mlp-sm_processed/2-mlp-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_mlp_out
    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
    153
    +0.14
    0.8%
    320
    +0.12
    0.7%
    492
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    255
    +0.14
    0.11
    492
    +0.12
    0.07
    331
    +0.11
    0.08
    Negative Logits
    cer
    -1.60
     importantly
    -1.53
     <
    -1.44
     *
    -1.43
    withstanding
    -1.41
    acular
    -1.39
     *"
    -1.36
     connective
    -1.36
     dated
    -1.35
    ment
    -1.32
    POSITIVE LOGITS
    heet
    1.93
    heets
    1.85
     themselves
    1.63
    forge
    1.60
    ource
    1.56
    ↵
    1.53
    1.53
    1.53
    ↵
    1.53
    1.53
    Activations Density 2.280%

    No Known Activations