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. 253
    Prev
    Next
    INDEX
    Explanations

    words related to sports terminology and events

    oai_token-act-pair · gpt-3.5-turbo

    terms related to freestyle performances and specific locations like Sochi

    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
    Winged
    -0.82
     Gard
    -0.71
     Surveillance
    -0.70
     newsletters
    -0.70
     Mages
    -0.69
     messenger
    -0.68
    Card
    -0.67
     blackmail
    -0.67
     gard
    -0.67
    letters
    -0.67
    POSITIVE LOGITS
    estyle
    2.76
     scrimmage
    1.89
     Sochi
    1.75
    estyles
    1.55
    inka
    1.50
    around
    1.34
    chery
    1.29
     simulate
    1.28
    rano
    1.22
    eco
    1.12
    Activations Density 0.050%

    No Known Activations