© 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 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. 2-RES-JB
    11. 5904
    Prev
    Next
    INDEX
    Explanations

    words related to technology systems, particularly those with high specificity or importance

    oai_token-act-pair · gpt-3.5-turbo

    references to various types of systems

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĤ¤ãĥĪ
    -0.78
    tein
    -0.77
    ishable
    -0.72
    past
    -0.70
     )]
    -0.69
    ãĥĥ
    -0.68
     earnest
    -0.68
    æĢ
    -0.66
    noon
    -0.65
    liest
    -0.64
    POSITIVE LOGITS
    atics
    0.98
     systems
    0.91
     integ
    0.87
     system
    0.86
    ycle
    0.81
    wide
    0.76
     tray
    0.75
    itially
    0.74
    ayne
    0.74
    ctl
    0.72
    Activations Density 0.042%

    No Known Activations