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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 4637
    Prev
    Next
    INDEX
    Explanations

    terms related to distinguishing or differentiating concepts or entities

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_8.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.8.hook_resid_post
    Hook Layer
    8
    Architecture
    standard
    Context Size
    64
    Dataset
    Skylion007/openwebtext
    Activation Function
    topk
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    679
    +0.12
    0.5%
    276
    +0.11
    0.5%
    288
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    276
    +0.12
    0.02
    382
    +0.11
    0.02
    679
    +0.10
    0.02
    Negative Logits
    vati
    -0.65
     Apostles
    -0.61
    CLA
    -0.61
     havoc
    -0.59
    nz
    -0.59
    conn
    -0.59
     Fulton
    -0.58
    MN
    -0.55
    elin
    -0.55
     Annotations
    -0.55
    POSITIVE LOGITS
     between
    1.18
    iates
    1.11
    iveness
    1.09
    ively
    1.00
     Between
    0.98
    iating
    0.95
    ially
    0.93
    between
    0.91
    iate
    0.87
    iator
    0.84
    Activations Density 0.060%

    No Known Activations