© 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. 2739
    Prev
    Next
    INDEX
    Explanations

    suspicious behavior or situations

    oai_token-act-pair · gpt-3.5-turbo

    references to suspicious behavior or activities

    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
    vation
    -0.71
    ode
    -0.68
    ffen
    -0.67
     Transition
    -0.65
    kamp
    -0.64
    tl
    -0.64
    elong
    -0.64
    Cost
    -0.64
     Dak
    -0.64
    tes
    -0.63
    POSITIVE LOGITS
     suspicious
    3.67
     suspic
    2.19
    susp
    1.72
     suspicions
    1.64
     suspected
    1.56
     suspect
    1.51
     dubious
    1.47
     suspicion
    1.46
     unexplained
    1.34
     paranoid
    1.31
    Activations Density 0.018%

    No Known Activations