© 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. 4785
    Prev
    Next
    INDEX
    Explanations

    ellipses or indications of omitted text

    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
    420
    +0.13
    0.6%
    755
    +0.11
    0.5%
    373
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    420
    +0.13
    0.01
    6
    +0.11
    0.01
    731
    +0.10
    0.01
    Negative Logits
    ttes
    -0.95
    etsk
    -0.83
    urses
    -0.77
     tremend
    -0.74
    cakes
    -0.71
    oded
    -0.70
    odies
    -0.70
    nces
    -0.70
    ning
    -0.70
    liness
    -0.68
    POSITIVE LOGITS
    ³³³³³³³³
    1.77
    ³³³³
    1.77
    ³³³³³³³³³³³³³³³³
    1.70
    ³³³
    1.67
    ³³
    1.63
     <!--
    0.90
    ↵Âł
    0.82
    -+
    0.76
    -|
    0.73
    ا
    0.72
    Activations Density 0.031%

    No Known Activations