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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsBlog/PodcastSlackPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlog/PodcastGitHubSlackTwitterContact
    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. 3275
    Prev
    Next
    INDEX
    Explanations

    references to various universities and institutions

    oai_token-act-pair · gpt-3.5-turbo

    the term "Sp" as well as its variations, indicating a focus on specific locations or contexts that start with those letters

    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
    aldehyde
    -0.79
    aho
    -0.74
    raq
    -0.70
    ailable
    -0.68
    enhagen
    -0.68
    atana
    -0.67
    etsu
    -0.66
     toget
    -0.66
    ãĤ©
    -0.66
    FUL
    -0.65
    POSITIVE LOGITS
     Sp
    3.45
    Sp
    2.34
     sp
    1.97
    sp
    1.91
     SP
    1.57
     Spr
    1.41
     Spin
    1.41
     Spy
    1.34
     Spe
    1.32
     Sph
    1.26
    Activations Density 0.025%

    No Known Activations