© 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. 5394
    Prev
    Next
    INDEX
    Explanations

    references to spatial directions or positions

    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
    408
    +0.12
    0.6%
    373
    +0.12
    0.5%
    678
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    596
    +0.12
    0.05
    435
    +0.12
    0.03
    99
    +0.10
    0.04
    Negative Logits
    ensibly
    -0.82
    onis
    -0.77
    arious
    -0.77
    perse
    -0.76
    olitics
    -0.75
    speak
    -0.75
    atically
    -0.75
    ãĤ¤ãĥĪ
    -0.74
    cially
    -0.73
    zynski
    -0.72
    POSITIVE LOGITS
     side
    1.15
    most
    1.03
     edge
    1.01
     corner
    0.99
     portion
    0.97
     underside
    0.96
     abdomen
    0.89
     hand
    0.89
     half
    0.89
     compartment
    0.88
    Activations Density 0.193%

    No Known Activations