© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 16451
    Prev
    Next
    INDEX
    Explanations

    words related to summoning or conjuring

    oai_token-act-pair · gpt-3.5-turbo
    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
    ificant
    -0.90
     PHI
    -0.82
    ratulations
    -0.80
    reements
    -0.78
     Confeder
    -0.77
    acters
    -0.76
    Interstitial
    -0.73
    backer
    -0.71
    ateral
    -0.71
     Savings
    -0.71
    POSITIVE LOGITS
     conj
    2.44
     summons
    1.89
     concoct
    1.87
     summoned
    1.74
     summon
    1.67
     pedd
    1.66
     unleashed
    1.66
     summoning
    1.64
     churn
    1.61
     Summon
    1.54
    Activations Density 0.043%

    No Known Activations