© 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. 3-RES-JB
    11. 23811
    Prev
    Next
    INDEX
    Explanations

    phrases and sentences that start with the word "If" and provide conditional instructions or information

    oai_token-act-pair · gpt-3.5-turbo

    conditional statements or clauses

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ãĤª
    -0.80
    stars
    -0.69
    FTWARE
    -0.69
    forth
    -0.68
    ãĤ¨ãĥ«
    -0.68
    earth
    -0.67
    ãĥŃ
    -0.66
    Ò
    -0.64
    ãĥ¬
    -0.64
    æ©
    -0.64
    POSITIVE LOGITS
    rame
    0.89
     you
    0.87
    uana
    0.86
    fy
    0.85
    soever
    0.84
    eret
    0.81
     applicable
    0.74
    yip
    0.70
     quer
    0.67
    cfg
    0.67
    Activations Density 0.115%

    No Known Activations