© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    5. GPT2-Small
    7. Residual Stream
    9. 7-RES-JB
    11. 15314
    Prev
    Next
    INDEX
    Explanations

    verbs or phrases indicating causing harm, damage, or negative impacts

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "cause" and its variations, emphasizing events or actions that lead to consequences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 7-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    7
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ian
    -0.72
     Technique
    -0.69
    atu
    -0.69
    stra
    -0.67
    ramid
    -0.66
     Niet
    -0.64
    halla
    -0.64
    schild
    -0.61
    framework
    -0.61
     skelet
    -0.61
    POSITIVE LOGITS
     havoc
    1.23
     headaches
    0.93
     mayhem
    0.91
    ãĥĨãĤ£
    0.91
     trouble
    0.91
     cele
    0.87
     irre
    0.84
     confusion
    0.83
     nightmares
    0.83
     unnecessary
    0.82
    Activations Density 0.055%

    No Known Activations