© 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. 13329
    Prev
    Next
    INDEX
    Explanations

    information or reports that have been released or published

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "released" indicating the publication of reports or data

    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
    isma
    -0.95
    iop
    -0.85
    anan
    -0.81
    stal
    -0.68
     impression
    -0.66
    throp
    -0.64
    tein
    -0.61
     Vij
    -0.60
    ourney
    -0.60
     Trou
    -0.60
    POSITIVE LOGITS
    released
    1.07
    æĸ¹
    0.93
     releasing
    0.92
     Releases
    0.91
    éĹĺ
    0.91
     RELEASE
    0.89
    quished
    0.84
     RELE
    0.83
     releases
    0.83
    Release
    0.83
    Activations Density 0.034%

    No Known Activations