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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 8009
    Prev
    Next
    INDEX
    Explanations

    words related to specific locations, particularly Middle Eastern cities and technologies

    oai_token-act-pair · gpt-3.5-turbo

    references to specific individuals or groups associated with the Islamic State

    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
    ttes
    -0.77
     Martial
    -0.71
     Merlin
    -0.67
    ective
    -0.66
     Trials
    -0.64
     Valkyrie
    -0.64
    -+
    -0.62
     Tune
    -0.62
     Inqu
    -0.61
    aur
    -0.61
    POSITIVE LOGITS
    aghd
    0.91
    hatt
    0.86
    istic
    0.74
    ishop
    0.73
    ingu
    0.73
    myra
    0.73
    \\\\\\\\
    0.72
    ifiable
    0.72
    aths
    0.72
    houn
    0.70
    Activations Density 0.020%

    No Known Activations