© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 28410
    Prev
    Next
    INDEX
    Explanations

    terms related to detailed analysis or descriptions of subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    373
    +0.15
    0.7%
    258
    +0.10
    0.4%
    140
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    159
    +0.15
    0.07
    158
    +0.10
    0.07
    122
    +0.10
    0.08
    Negative Logits
    cffff
    -0.69
    TPS
    -0.68
     Cancel
    -0.68
    locked
    -0.67
    ickle
    -0.65
     Clever
    -0.65
    Variable
    -0.64
    LINE
    -0.64
    âĻ¥
    -0.63
    channelAvailability
    -0.62
    POSITIVE LOGITS
     glimpse
    1.40
     account
    1.39
     portrayal
    1.35
     depiction
    1.31
     look
    1.31
     overview
    1.30
     explanation
    1.24
     view
    1.23
     description
    1.19
     portrait
    1.18
    Activations Density 1.735%

    No Known Activations