© 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. 23898
    Prev
    Next
    INDEX
    Explanations

    phrases related to moral judgment, reprehensible actions, and actions that are morally questionable

    oai_token-act-pair · gpt-3.5-turbo

    negative moral judgments or the concept of moral reprehensibility

    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
     throats
    -0.86
    gdala
    -0.77
     packs
    -0.77
     cohorts
    -0.69
     assemblies
    -0.67
     crowds
    -0.67
     flock
    -0.65
    deen
    -0.65
     contacts
    -0.65
    */(
    -0.65
    POSITIVE LOGITS
    ceivable
    1.36
    ivable
    1.32
    usable
    1.26
    icable
    1.20
    atable
    1.18
    vable
    1.17
    urable
    1.16
    risome
    1.12
    ailable
    1.10
    orable
    1.09
    Activations Density 0.213%

    No Known Activations