© 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. 4986
    Prev
    Next
    INDEX
    Explanations

    Instances of the word "overt" and related phrases indicating explicitness or directness

    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
    127
    +0.11
    0.5%
    314
    +0.10
    0.4%
    187
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    406
    +0.11
    0.04
    326
    +0.10
    0.04
    314
    +0.09
    -0.00
    Negative Logits
    ity
    -0.75
    alogy
    -0.67
    fighter
    -0.66
    istries
    -0.63
    acity
    -0.63
     Cyborg
    -0.62
    itor
    -0.62
    istry
    -0.61
    itous
    -0.61
    uality
    -0.59
    POSITIVE LOGITS
    asonic
    0.80
     redes
    0.79
     teasp
    0.78
    hesda
    0.76
    ¿½
    0.76
    aking
    0.75
    eering
    0.74
    aken
    0.74
    robe
    0.74
    -+-+
    0.71
    Activations Density 0.233%

    No Known Activations