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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 0-RES-JB
    11. 15277
    Prev
    Next
    INDEX
    Explanations

    expressions signaling blame or responsibility for something negative

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of blame

    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
    aza
    -0.68
     Unique
    -0.66
     Rounds
    -0.65
    Champ
    -0.63
     Conver
    -0.60
     Powered
    -0.59
    col
    -0.59
    Fin
    -0.59
     iss
    -0.58
     SPR
    -0.57
    POSITIVE LOGITS
     blame
    4.42
     blaming
    2.17
     blames
    2.15
     blamed
    2.00
     fault
    1.85
     scapego
    1.82
     culp
    1.56
     responsibility
    1.52
     guilt
    1.37
     accuse
    1.34
    Activations Density 0.009%

    No Known Activations