© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 11959
    Prev
    Next
    INDEX
    Explanations

    words related to endings, conclusions, or final parts of something

    oai_token-act-pair · gpt-3.5-turbo

    references to endings or conclusions in various contexts

    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
    hee
    -0.87
    inx
    -0.83
    itsch
    -0.80
    otta
    -0.76
    icka
    -0.72
    ña
    -0.69
    icipated
    -0.68
    enko
    -0.68
    é¾įå¥ij士
    -0.68
    committee
    -0.68
    POSITIVE LOGITS
     endings
    0.84
    angering
    0.84
     tragically
    0.77
    doors
    0.75
     prematurely
    0.73
    ocrine
    0.70
    owment
    0.69
     sidel
    0.68
     ending
    0.67
    tackle
    0.67
    Activations Density 0.014%

    No Known Activations