© 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. Attention Out - 32k
    9. 3-ATT_32K-OAI
    11. 4021
    Prev
    Next
    INDEX
    Explanations

    references to evaluations and analyses, particularly in relation to social programs and policies

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-att_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-attn-out-SAEs/v5_32k_layer_3
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.3.hook_attn_out
    Hook Layer
    3
    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
    481
    +0.10
    0.4%
    53
    +0.09
    0.4%
    713
    +0.09
    0.4%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    321
    +0.10
    0.07
    358
    +0.09
    0.07
    444
    +0.09
    0.06
    Negative Logits
    fal
    -0.80
    fty
    -0.66
    rily
    -0.64
    umber
    -0.61
    strom
    -0.61
     Studios
    -0.61
    raught
    -0.60
    llor
    -0.60
     olive
    -0.60
    esters
    -0.60
    POSITIVE LOGITS
     Pengu
    0.82
     threats
    0.71
    itability
    0.69
     propensity
    0.69
     persistence
    0.68
     potential
    0.65
    itiveness
    0.64
     pets
    0.64
    overe
    0.64
    oneliness
    0.64
    Activations Density 1.449%

    No Known Activations