© 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. 17223
    Prev
    Next
    INDEX
    Explanations

    family relationships, particularly between fathers and sons

    oai_token-act-pair · gpt-3.5-turbo

    references to parental figures, particularly fathers and mothers

    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
     Flavoring
    -0.91
    âĶĢâĶĢâĶĢâĶĢ
    -0.71
    ormon
    -0.67
    psey
    -0.64
    apo
    -0.64
    resso
    -0.63
    âĶĢâĶĢ
    -0.63
    vernment
    -0.62
    ocobo
    -0.61
    leans
    -0.61
    POSITIVE LOGITS
    hetical
    0.89
    hetically
    0.84
    hesis
    0.80
    heses
    0.80
    hood
    0.79
    ples
    0.67
    bang
    0.67
     childbirth
    0.67
    ship
    0.67
    nect
    0.67
    Activations Density 0.043%

    No Known Activations