© 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. 15271
    Prev
    Next
    INDEX
    Explanations

    attributes related to physical abilities or skills, such as strength and dexterity

    oai_token-act-pair · gpt-3.5-turbo

    references to attributes and characteristics in games or role-playing contexts

    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
    rador
    -0.89
    schild
    -0.73
    terday
    -0.72
     targ
    -0.70
     hosp
    -0.69
     harbor
    -0.68
     harbour
    -0.67
     nu
    -0.66
     mural
    -0.66
    papers
    -0.64
    POSITIVE LOGITS
     Strength
    2.84
     Dexterity
    1.76
     Hearts
    1.64
    Strength
    1.57
     Charisma
    1.43
     Agility
    1.27
     Wisdom
    1.19
    strength
    1.10
    ogle
    0.92
     Mastery
    0.86
    Activations Density 0.016%

    No Known Activations