© 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. 4999
    Prev
    Next
    INDEX
    Explanations

    references to religious institutions and community organizations

    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
    712
    +0.12
    0.6%
    267
    +0.12
    0.5%
    224
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    146
    +0.12
    0.08
    288
    +0.12
    0.10
    267
    +0.10
    0.08
    Negative Logits
     strokes
    -0.80
     embodiments
    -0.76
    ometric
    -0.72
    liga
    -0.69
     Frieza
    -0.66
    tools
    -0.65
     instruments
    -0.64
     clauses
    -0.64
     stroke
    -0.64
     measures
    -0.63
    POSITIVE LOGITS
     Ltd
    1.03
     located
    0.89
     LLC
    0.85
    astery
    0.81
     Inc
    0.80
     LLP
    0.80
     LTD
    0.76
     Limited
    0.76
     Parkway
    0.76
     Scarborough
    0.76
    Activations Density 0.705%

    No Known Activations