© 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. 10-RES-JB
    11. 20388
    Prev
    Next
    INDEX
    Explanations

    phrases containing names of well-known individuals

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of specific letter combinations or sequences within words

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 10-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    10
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    BIP
    -0.69
    channelAvailability
    -0.69
     CPC
    -0.68
     Chapman
    -0.57
     separat
    -0.55
    anchester
    -0.54
    shaw
    -0.54
     confines
    -0.53
     IPM
    -0.52
    SHIP
    -0.52
    POSITIVE LOGITS
    Wan
    0.93
    uary
    0.92
    leck
    0.85
    ratulations
    0.82
    orio
    0.82
    ument
    0.79
    worldly
    0.78
    wald
    0.76
    yssey
    0.75
    actory
    0.71
    Activations Density 0.093%

    No Known Activations