© 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. 9-RES-JB
    11. 1000
    Prev
    Next
    INDEX
    Explanations

    references to legal or criminal activities

    oai_token-act-pair · gpt-3.5-turbo

    references to criminal intent or involvement

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    Cooldown
    -0.74
    gravity
    -0.72
    ursday
    -0.72
    population
    -0.71
    estival
    -0.71
     brunch
    -0.70
    pop
    -0.70
    byss
    -0.69
     Xander
    -0.68
    grim
    -0.67
    POSITIVE LOGITS
     culp
    1.47
     wrongdoing
    1.32
     complicit
    1.32
     knowingly
    1.28
     complicity
    1.22
     unlawfully
    1.21
     misconduct
    1.16
     collusion
    1.15
     negligence
    1.13
     tampering
    1.13
    Activations Density 0.624%

    No Known Activations