© 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. 11-RES-JB
    11. 2686
    Prev
    Next
    INDEX
    Explanations

    references to demonic or negative connotations, often associated with protests or criticisms

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "demonization" in any context

    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
     Morales
    -0.70
    opher
    -0.59
    âĢİ
    -0.59
    ippi
    -0.59
    abouts
    -0.59
    IGH
    -0.58
     Fahrenheit
    -0.58
    arella
    -0.57
     Greenberg
    -0.57
    orney
    -0.57
    POSITIVE LOGITS
    stration
    1.80
    strate
    1.54
    str
    1.40
    iac
    1.07
    ization
    0.99
    stru
    0.95
    oid
    0.95
    ufact
    0.94
    isation
    0.91
    et
    0.90
    Activations Density 0.025%

    No Known Activations