© 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. 0-RES-JB
    11. 15215
    Prev
    Next
    INDEX
    Explanations

    superlatives, particularly those related to positive attributes

    oai_token-act-pair · gpt-3.5-turbo

    superlative forms of adjectives

    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
     embedded
    -0.58
     secondary
    -0.57
     General
    -0.56
     fac
    -0.56
     shock
    -0.55
     such
    -0.55
     pseud
    -0.55
     applied
    -0.54
     parallel
    -0.54
     persons
    -0.54
    POSITIVE LOGITS
    iest
    4.46
    liest
    2.37
    iness
    2.32
    ier
    2.22
    ily
    1.98
    hest
    1.69
    lest
    1.47
    ieth
    1.47
    iers
    1.43
    est
    1.41
    Activations Density 0.008%

    No Known Activations