© 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. 3038
    Prev
    Next
    INDEX
    Explanations

    comparative terms and concepts indicating superiority or inferiority

    oai_token-act-pair · gpt-3.5-turbo

    references to superiority and inferiority among groups or entities

    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
    BRE
    -0.92
    mberg
    -0.88
     Dresden
    -0.87
    gren
    -0.79
    RPG
    -0.78
    MG
    -0.78
    SN
    -0.74
    mitt
    -0.73
    è¦ļéĨĴ
    -0.73
    OWS
    -0.71
    POSITIVE LOGITS
     superior
    0.89
     dilig
    0.86
     inferior
    0.82
     grades
    0.78
     educ
    0.77
    intendent
    0.76
     intellect
    0.75
     judgement
    0.72
     nomine
    0.72
    iour
    0.72
    Activations Density 0.010%

    No Known Activations