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. 11-RES-JB
    6. 9770
    Prev
    Next
    INDEX
    Explanations

    technical and industry-related phrases and terms

    oai_token-act-pair · gpt-3.5-turbo

    descriptive terms related to significant concepts or events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     unbearable
    -0.66
     intolerable
    -0.65
     uncontroll
    -0.61
     Âł Âł
    -0.58
     humiliating
    -0.58
    ,,,,
    -0.57
    ··
    -0.57
     etc
    -0.56
     improper
    -0.55
     humili
    -0.55
    POSITIVE LOGITS
     Kickstarter
    0.71
     specs
    0.68
     pioneered
    0.68
     modular
    0.68
    branded
    0.67
     innovation
    0.67
     prototype
    0.67
    itect
    0.66
    gorithm
    0.66
     futuristic
    0.65
    Activations Density 0.727%

    No Known Activations