© 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
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 0-RES-JB
    6. 19140
    Prev
    Next
    INDEX
    Explanations

    phrases related to instructions, procedures, or formal statements

    oai_token-act-pair · gpt-3.5-turbo

    mentions of "owl."

    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
     Atlantis
    -0.73
     sul
    -0.68
     Nik
    -0.64
     Britann
    -0.62
     Tas
    -0.61
     mir
    -0.60
     spir
    -0.59
     Monica
    -0.59
     none
    -0.58
     Meteor
    -0.58
    POSITIVE LOGITS
    OW
    4.22
    OWS
    2.86
    ows
    2.24
    owed
    1.92
    ow
    1.89
    owing
    1.78
    OWER
    1.72
    OWN
    1.52
    ower
    1.47
    OU
    1.40
    Activations Density 0.004%

    No Known Activations