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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 3-RES-JB
    11. 544
    Prev
    Next
    INDEX
    Explanations

    mentions of organizational or governmental bodies

    oai_token-act-pair · gpt-3.5-turbo

    references to various organizational entities or groups

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Kafka
    -0.75
     Job
    -0.68
     Dickens
    -0.68
     Hoover
    -0.67
     desperation
    -0.65
    edia
    -0.62
    kers
    -0.62
    Zip
    -0.59
    ophobic
    -0.57
     Typh
    -0.57
    POSITIVE LOGITS
    guards
    1.11
    guard
    0.90
    builders
    0.87
    anguage
    0.85
    lain
    0.81
    building
    0.79
    weight
    0.79
    chair
    0.78
     politic
    0.77
    shed
    0.76
    Activations Density 0.017%

    No Known Activations