Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 5236
    Prev
    Next
    INDEX
    Explanations

    words related to reproduction or introduction

    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
    0
    +0.13
    0.6%
    746
    +0.11
    0.5%
    98
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    555
    +0.13
    0.03
    746
    +0.11
    0.02
    232
    +0.09
    0.03
    Negative Logits
    rieve
    -0.70
     Frie
    -0.66
    arta
    -0.65
    room
    -0.65
     Immunity
    -0.64
    uyomi
    -0.64
    RAW
    -0.62
    AIN
    -0.61
    oola
    -0.59
    ARB
    -0.58
    POSITIVE LOGITS
    cing
    1.88
    ced
    1.70
    cer
    1.61
    ctions
    1.60
    cers
    1.60
    ctory
    1.60
    ces
    1.54
    ctive
    1.49
    cence
    1.45
    ce
    1.39
    Activations Density 0.137%

    No Known Activations