© 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. 572
    Prev
    Next
    INDEX
    Explanations

    phrases or sentences referring to processes or actions that may lead to significant changes or impacts

    oai_token-act-pair · gpt-3.5-turbo

    phrases and contexts that discuss the concept of "way" or methods of doing things

    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
    usters
    -0.98
    uster
    -0.94
    encer
    -0.77
     livest
    -0.74
     sugg
    -0.71
    erville
    -0.70
    igh
    -0.68
    acca
    -0.68
    ĸļ
    -0.67
     subsequ
    -0.67
    POSITIVE LOGITS
    fare
    0.98
    ward
    0.89
    finding
    0.87
    point
    0.78
    bill
    0.75
     WAY
    0.70
    lay
    0.69
    sey
    0.67
    forward
    0.65
    points
    0.64
    Activations Density 0.039%

    No Known Activations