© 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 EvalsExportsinterp-engineNEWAPI 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. 12-RES-JB
    11. 4423
    Prev
    Next
    INDEX
    Explanations

    descriptive terms related to artistic or literary work, as well as quantities and entities

    oai_token-act-pair · gpt-3.5-turboTriggered by @bot

    references to the concept of "sheer" magnitude or intensity in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 12-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post
    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_post
    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
     Tags
    -0.75
    ublic
    -0.71
    letters
    -0.69
     Handler
    -0.69
     Bye
    -0.67
    arget
    -0.65
    ãĥ¼ãĥ³
    -0.65
    pai
    -0.64
    taker
    -0.63
     Caption
    -0.61
    POSITIVE LOGITS
     brilliance
    1.52
     majesty
    1.51
     splend
    1.44
     sophistication
    1.39
     ingenuity
    1.38
     complexity
    1.37
     beauty
    1.37
     enorm
    1.36
     versatility
    1.35
     richness
    1.34
    Activations Density 0.371%

    No Known Activations