Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 16217
    Prev
    Next
    INDEX
    Explanations

    Vietnam War-related terms, possibly focusing on key battles or events

    oai_token-act-pair · gpt-3.5-turbo

    references to various series or battles in the Gundam franchise

    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
    mop
    -0.77
    911
    -0.76
    owell
    -0.75
    prim
    -0.72
     Idaho
    -0.70
    rib
    -0.70
    lake
    -0.69
    quart
    -0.68
    Pal
    -0.68
     Kuwait
    -0.67
    POSITIVE LOGITS
     VS
    3.35
     Vs
    1.58
     Grad
    1.15
    ilda
    1.13
     Chronicles
    1.10
     Versus
    1.06
     XD
    1.05
     Vik
    0.99
     Remix
    0.96
     Dracula
    0.93
    Activations Density 0.037%

    No Known Activations