© 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. 3-RES-JB
    11. 1774
    Prev
    Next
    INDEX
    Explanations

    phrases related to the making or creation of things

    oai_token-act-pair · gpt-3.5-turbo

    gerund forms of verbs suggesting active processes or actions

    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
    istration
    -0.88
    eded
    -0.77
    è£
    -0.73
    к
    -0.71
    imposed
    -0.70
    umbn
    -0.70
    оÐ
    -0.70
    pent
    -0.69
    ÑĢ
    -0.68
    ulz
    -0.68
    POSITIVE LOGITS
     Ones
    1.02
     Them
    1.00
     Gets
    0.98
     Comes
    0.97
     Makes
    0.97
     Difference
    0.96
     Advice
    0.95
     Wrong
    0.95
     Names
    0.95
     Together
    0.95
    Activations Density 0.168%

    No Known Activations