Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 11187
    Prev
    Next
    INDEX
    Explanations

    strong, persistent, and intense characteristics or actions

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "relentless" and its variations

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    fax
    -0.72
     differs
    -0.68
    alde
    -0.67
    Options
    -0.63
    ook
    -0.62
    zech
    -0.62
    qua
    -0.61
    icka
    -0.61
    ippi
    -0.61
     items
    -0.61
    POSITIVE LOGITS
     relentless
    2.96
     relentlessly
    2.01
     unrel
    2.01
     incess
    1.97
     continual
    1.81
     constant
    1.75
     perpetual
    1.70
     persistent
    1.50
     merciless
    1.49
     dogged
    1.48
    Activations Density 0.038%

    No Known Activations