© 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
    2. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    3. Pythia-70M-Deduped
    4. Residuals
    5. 2-RES-SM
    6. 28753
    Prev
    Next
    INDEX
    Explanations

    technical terms and components related to materials and manufacturing processes

    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
    231
    +0.19
    1.1%
    47
    +0.14
    0.8%
    204
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    231
    +0.19
    0.08
    47
    +0.14
    0.09
    204
    +0.12
    0.07
    Negative Logits
     alert
    -2.02
     clock
    -1.76
     longitude
    -1.74
     sleep
    -1.61
     briefing
    -1.60
    eor
    -1.55
     mood
    -1.55
     dispat
    -1.54
     conference
    -1.54
    Sleep
    -1.54
    POSITIVE LOGITS
    ľ
    2.88
    ¼
    2.85
    Ļ
    2.68
    »¿
    2.53
    ·¸
    2.49
    µ
    2.45
    IJ
    2.38
    ħ
    2.37
    Ļª
    2.37
    Ķ
    2.33
    Activations Density 0.981%

    No Known Activations