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. 7387
    Prev
    Next
    INDEX
    Explanations

    text related to backing up data or actions

    oai_token-act-pair · gpt-3.5-turbo

    phrases and concepts related to backup and recovery systems

    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
    STEM
    -0.80
    merce
    -0.70
    Weather
    -0.70
    vic
    -0.70
    Plot
    -0.69
    LGBT
    -0.69
    Deploy
    -0.67
    igham
    -0.67
    »Ĵ
    -0.66
    Specific
    -0.65
    POSITIVE LOGITS
    zig
    0.73
     Tide
    0.71
     whence
    0.69
     sofa
    0.68
     Forth
    0.66
     haunt
    0.65
     Gret
    0.65
     corners
    0.64
     forth
    0.62
     vengeance
    0.62
    Activations Density 0.244%

    No Known Activations