© 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. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 8-RES_POST_32K-OAI
    11. 5415
    Prev
    Next
    INDEX
    Explanations

    instances of significant numerical or statistical data points

    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
    102
    +0.11
    0.5%
    186
    +0.10
    0.5%
    149
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    160
    +0.11
    0.11
    640
    +0.10
    0.10
    442
    +0.09
    0.10
    Negative Logits
     XII
    -0.71
     XIII
    -0.67
     Lv
    -0.66
     Directory
    -0.66
    OUS
    -0.66
     VIII
    -0.65
     Axis
    -0.64
     VI
    -0.62
     II
    -0.61
     XIV
    -0.61
    POSITIVE LOGITS
    instead
    1.06
    without
    1.03
    etheless
    1.03
    according
    0.95
    that
    0.94
    having
    0.92
    they
    0.92
    although
    0.91
    until
    0.91
    despite
    0.91
    Activations Density 0.331%

    No Known Activations