© 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. 2654
    Prev
    Next
    INDEX
    Explanations

    references to speakers, often related to events or organizations

    oai_token-act-pair · gpt-3.5-turbo

    references to speakers and speaking engagements

    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
     Soldiers
    -0.71
    urat
    -0.70
     Devils
    -0.69
     Mines
    -0.68
    hens
    -0.66
     Pradesh
    -0.62
    laus
    -0.61
    uilt
    -0.61
    ster
    -0.59
    aredevil
    -0.58
    POSITIVE LOGITS
     speaker
    0.97
    phone
    0.95
     speakers
    0.95
    glass
    0.88
    loo
    0.83
    hip
    0.82
    parts
    0.82
    boys
    0.79
    ofer
    0.79
    voice
    0.79
    Activations Density 0.036%

    No Known Activations