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. 3-RES-JB
    6. 11825
    Prev
    Next
    INDEX
    Explanations

    terms related to prototypes

    oai_token-act-pair · gpt-3.5-turbo

    references to "proto" entities or terms within a certain context

    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
    ADRA
    -0.86
    EngineDebug
    -0.76
    ======
    -0.76
     Learns
    -0.76
     Fargo
    -0.73
     [|
    -0.73
    KING
    -0.73
    HCR
    -0.72
    BIP
    -0.71
    BOOK
    -0.71
    POSITIVE LOGITS
    ractor
    0.96
     prot
    0.96
    inent
    0.95
    uber
    0.94
    ract
    0.85
    ocol
    0.83
    etrical
    0.83
    etric
    0.82
    otyp
    0.81
    racted
    0.80
    Activations Density 0.013%

    No Known Activations