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. 8-RES-JB
    6. 16106
    Prev
    Next
    INDEX
    Explanations

    words that express blame or responsibility

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "blame" and its related forms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 8-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    8
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    intend
    -0.81
    quire
    -0.72
    tein
    -0.72
    chedel
    -0.69
    weeney
    -0.68
    cephal
    -0.68
    ierre
    -0.67
    mental
    -0.66
    naissance
    -0.66
    ser
    -0.65
    POSITIVE LOGITS
     blame
    1.15
     scapego
    0.95
     blames
    0.90
     blaming
    0.90
     squarely
    0.86
     blamed
    0.81
     victim
    0.80
     culprit
    0.79
     obsc
    0.72
     victims
    0.71
    Activations Density 0.046%

    No Known Activations