© 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. 11-RES-JB
    11. 20699
    Prev
    Next
    INDEX
    Explanations

    audio-related terms

    oai_token-act-pair · gpt-3.5-turbo

    references to audio content and recordings

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 11-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    11
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ary
    -0.88
    ndra
    -0.82
    lon
    -0.79
    nel
    -0.78
    rir
    -0.76
    laws
    -0.76
    ality
    -0.76
    vier
    -0.75
    naire
    -0.74
    redo
    -0.72
    POSITIVE LOGITS
     recordings
    1.18
     cassette
    1.12
     recording
    1.10
     playback
    1.08
     codec
    1.07
     recorder
    1.04
     cass
    0.97
    obook
    0.96
     narration
    0.92
     audio
    0.90
    Activations Density 0.035%

    No Known Activations