© 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. 4864
    Prev
    Next
    INDEX
    Explanations

    key terms related to organized groups or teams, particularly in a competitive or collaborative context

    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
    13
    +0.10
    0.5%
    352
    +0.10
    0.5%
    7
    +0.10
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.10
    0.01
    40
    +0.10
    0.01
    136
    +0.10
    0.01
    Negative Logits
    omics
    -0.74
    iber
    -0.71
    htaking
    -0.70
    uses
    -0.68
    lli
    -0.67
    issance
    -0.67
    access
    -0.65
    nesia
    -0.64
    istries
    -0.63
    clair
    -0.62
    POSITIVE LOGITS
    WARD
    0.81
     Skydragon
    0.80
     Cowboy
    0.72
    INGTON
    0.69
    bour
    0.69
     SQU
    0.69
     GOODMAN
    0.68
     bour
    0.66
    acco
    0.64
    NER
    0.64
    Activations Density 0.009%

    No Known Activations