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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports 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. 8692
    Prev
    Next
    INDEX
    Explanations

    phrases related to the musician "Weezer" specifically when mentioning their lead vocalist, Rivers Cuomo

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the substring 'ez' in words

    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
     IGF
    -0.69
     Institutes
    -0.62
    photos
    -0.61
    ISS
    -0.60
     Scandinav
    -0.60
     Icelandic
    -0.59
     glim
    -0.59
    OGR
    -0.59
    izations
    -0.58
     Malays
    -0.58
    POSITIVE LOGITS
    hner
    1.00
    ombie
    0.95
    hou
    0.95
    hov
    0.93
    hao
    0.90
    cellent
    0.90
    terday
    0.89
    ual
    0.87
    cue
    0.87
    illas
    0.85
    Activations Density 0.035%

    No Known Activations