Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 11-RES-JB
    6. 20515
    Prev
    Next
    INDEX
    Explanations

    references to specific titles or names, likely related to entertainment media such as movies, music, and literature

    oai_token-act-pair · gpt-3.5-turbo

    references to music or bands

    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
     credential
    -0.84
     allowable
    -0.79
     elector
    -0.77
     elig
    -0.75
     AFP
    -0.73
     oblig
    -0.73
     hor
    -0.73
     unnecessarily
    -0.72
     complementary
    -0.72
     affordable
    -0.72
    POSITIVE LOGITS
     Rider
    1.18
    bringer
    1.16
    punk
    1.16
     Girl
    1.13
    Girl
    1.13
    breaker
    1.10
     Breaker
    1.10
     Rats
    1.09
     Eater
    1.09
     Racer
    1.08
    Activations Density 0.254%

    No Known Activations