© 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
    2. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    3. GPT2-Small
    4. Residual Post - 32k
    5. 8-RES_POST_32K-OAI
    6. 4906
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "significant."

    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
    579
    +0.11
    0.5%
    623
    +0.10
    0.5%
    316
    +0.10
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    661
    +0.11
    0.03
    316
    +0.10
    0.03
    623
    +0.10
    0.03
    Negative Logits
    ©¶æ
    -0.97
    nery
    -0.78
    iverpool
    -0.74
    yer
    -0.69
    creen
    -0.68
    elf
    -0.67
    eez
    -0.66
    tein
    -0.66
     cz
    -0.64
    bable
    -0.64
    POSITIVE LOGITS
    itarian
    1.21
     amounts
    1.11
     amount
    1.01
     quantities
    0.92
     portions
    0.90
     understatement
    0.85
     imped
    0.83
     depressive
    0.82
     accomplishment
    0.82
     differences
    0.81
    Activations Density 0.078%

    No Known Activations