© 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. 1-RES-JB
    11. 20050
    Prev
    Next
    INDEX
    Explanations

    names of people or places containing the string "Kaw"

    oai_token-act-pair · gpt-3.5-turbo

    names of individuals or entities, particularly those related to specific cultural or entertainment contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Scotia
    -0.70
     Murdoch
    -0.63
     Malfoy
    -0.62
     discriminate
    -0.62
    Failure
    -0.61
    bite
    -0.60
    ional
    -0.60
     chemotherapy
    -0.60
     confinement
    -0.60
     purge
    -0.60
    POSITIVE LOGITS
     Kaw
    1.42
    asaki
    1.12
    aii
    1.04
    arij
    0.93
     Fuk
    0.90
    oji
    0.89
    eah
    0.87
    oru
    0.87
    apon
    0.85
    oka
    0.85
    Activations Density 0.011%

    No Known Activations