© 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. 4-RES-JB
    11. 20651
    Prev
    Next
    INDEX
    Explanations

    mentions of international politics and military actions

    oai_token-act-pair · gpt-3.5-turbo

    references to specific locations or entities associated with significant events or actions

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

    No Comments

    Negative Logits
    ĪĴ
    -0.53
     Leilan
    -0.50
    ãĤ©
    -0.47
    Lago
    -0.45
    }.
    -0.44
     Emin
    -0.44
     Carbuncle
    -0.44
    ãĥ¼ãĥĨ
    -0.43
     evolves
    -0.43
    Winged
    -0.43
    POSITIVE LOGITS
    ishly
    0.43
    arming
    0.40
     immunity
    0.40
    matically
    0.40
     appeal
    0.39
    iband
    0.39
    mods
    0.38
    helps
    0.37
    wide
    0.37
     perks
    0.37
    Activations Density 3.158%

    No Known Activations