© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. OpenAI · Sparse Autoencoder for GPT2-Small - v5 2024
    5. GPT2-Small
    7. Residual Post - 32k
    9. 10-RES_POST_32K-OAI
    11. 29055
    Prev
    Next
    INDEX
    Explanations

    statistical data or performance metrics related to basketball players

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res_post_32k-oai
    Configuration
    jbloom/GPT2-Small-OAI-v5-32k-resid-post-SAEs/v5_32k_layer_10.pt
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.10.hook_resid_post
    Hook Layer
    10
    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
    411
    +0.12
    0.5%
    203
    +0.11
    0.5%
    626
    +0.11
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    479
    +0.12
    0.02
    417
    +0.11
    0.06
    669
    +0.11
    0.08
    Negative Logits
     Sv
    -0.81
     Spark
    -0.78
     EG
    -0.78
     Aus
    -0.77
    iris
    -0.76
     Sig
    -0.75
    esp
    -0.75
     Keane
    -0.75
     AIR
    -0.74
    OIL
    -0.74
    POSITIVE LOGITS
     Mon
    2.50
    Mon
    2.34
    mon
    2.09
     mon
    2.03
     MON
    1.87
    MON
    1.63
     Mono
    1.60
     mono
    1.42
     Monk
    1.40
     Monroe
    1.35
    Activations Density 5.846%

    No Known Activations