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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 4-RES-JB
    11. 2320
    Prev
    Next
    INDEX
    Explanations

    references to a specific overall draft position

    oai_token-act-pair · gpt-3.5-turbo

    mentions of draft picks and their overall rankings in sports contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 4-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    24,576
    Data Type
    torch.float32
    Hook Point
    blocks.4.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    4
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Hyper
    -0.73
    ctions
    -0.71
     Speedway
    -0.68
     eld
    -0.67
    =/
    -0.67
    mosp
    -0.67
    Ly
    -0.66
    heter
    -0.66
     Witnesses
    -0.65
    cop
    -0.65
    POSITIVE LOGITS
     standings
    0.87
     constitu
    0.71
     landslide
    0.71
    heastern
    0.70
     rankings
    0.70
     qualifier
    0.70
     legalization
    0.70
     nationally
    0.69
     legalizing
    0.69
     hurdle
    0.68
    Activations Density 0.011%

    No Known Activations