© 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. 0-RES-JB
    11. 11551
    Prev
    Next
    INDEX
    Explanations

    phrases related to moral judgment or opinion

    oai_token-act-pair · gpt-3.5-turbo

    references to the concept of "thing" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 0-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    0
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     Cosponsors
    -0.58
    ango
    -0.57
     ranging
    -0.57
     Courts
    -0.56
    align
    -0.56
     thereby
    -0.56
     thus
    -0.56
    its
    -0.55
     Centers
    -0.53
     Forced
    -0.53
    POSITIVE LOGITS
     thing
    3.60
    thing
    2.15
     Thing
    2.11
     things
    1.64
    THING
    1.49
     stuff
    1.39
    things
    1.31
     guy
    1.20
     damn
    1.19
     Things
    1.19
    Activations Density 0.032%

    No Known Activations