© 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. 31016
    Prev
    Next
    INDEX
    Explanations

    specific names and terms related to film industry revenues and genres

    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
    +0.12
    0.5%
    191
    +0.10
    0.4%
    688
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    554
    +0.12
    -0.00
    98
    +0.10
    0.01
    468
    +0.10
    0.01
    Negative Logits
     247
    -0.83
     152
    -0.82
     248
    -0.82
     423
    -0.81
     336
    -0.79
     144
    -0.79
     232
    -0.78
     323
    -0.77
     324
    -0.77
     153
    -0.77
    POSITIVE LOGITS
    language
    0.86
    avers
    0.81
    directed
    0.81
    dayName
    0.78
    gon
    0.76
    amar
    0.76
    Pacific
    0.75
    phas
    0.74
    LOS
    0.73
    lined
    0.73
    Activations Density 0.064%

    No Known Activations