© 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. 9-RES-JB
    11. 4795
    Prev
    Next
    INDEX
    Explanations

    mentions of individuals being convicted of crimes

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "convicted."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 9-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    9
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ffee
    -0.68
    mentation
    -0.68
    eah
    -0.67
    edia
    -0.66
    mented
    -0.63
    ophe
    -0.63
     Remastered
    -0.62
    idge
    -0.62
    gm
    -0.62
    autions
    -0.61
    POSITIVE LOGITS
     felon
    1.16
     guilty
    0.96
     fel
    0.93
     unfocusedRange
    0.93
     convict
    0.90
     sentenced
    0.81
     convicted
    0.80
     sentencing
    0.79
     manslaughter
    0.78
    eering
    0.78
    Activations Density 0.036%

    No Known Activations