© 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. 1-RES-JB
    11. 14099
    Prev
    Next
    INDEX
    Explanations

    references to the music and entertainment network MTV

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the music television network MTV and related artists

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 1-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    1
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    erb
    -0.69
    htaking
    -0.67
    half
    -0.66
    thinking
    -0.64
    eem
    -0.63
    hiro
    -0.61
    +++
    -0.61
     healer
    -0.61
    ients
    -0.60
    lean
    -0.60
    POSITIVE LOGITS
     MTV
    1.11
     Sparkle
    0.73
    aji
    0.70
     Idol
    0.69
     contrace
    0.68
    angelo
    0.68
    ingle
    0.68
     Nadu
    0.67
    ortmund
    0.67
    illion
    0.66
    Activations Density 0.006%

    No Known Activations