© 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. 5389
    Prev
    Next
    INDEX
    Explanations

    occurrences of the letter 'H' and related characters, often in conjunction with specific names or actions

    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
    207
    +0.10
    0.5%
    313
    +0.10
    0.5%
    121
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    659
    +0.10
    0.07
    314
    +0.10
    0.01
    176
    +0.09
    0.04
    Negative Logits
    mbuds
    -0.83
    iosyncr
    -0.77
    ingers
    -0.68
    lik
    -0.67
    aneers
    -0.67
    rollers
    -0.65
    minster
    -0.64
    udeau
    -0.64
    metics
    -0.64
    este
    -0.63
    POSITIVE LOGITS
    POST
    0.68
    GET
    0.67
    :\
    0.67
    nil
    0.66
    Yo
    0.64
    wcs
    0.64
    Drop
    0.64
    DE
    0.64
    Hy
    0.63
    Users
    0.63
    Activations Density 0.454%

    No Known Activations