© 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. 2-RES-JB
    11. 23159
    Prev
    Next
    INDEX
    Explanations

    the word "Moss" with high activation values

    oai_token-act-pair · gpt-3.5-turbo

    mentions of the term "Moss" and its variations in the context of various subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 2-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    2
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    versions
    -0.75
    joined
    -0.69
    cers
    -0.67
    ACTED
    -0.65
    HCR
    -0.64
    IENCE
    -0.64
    Interstitial
    -0.64
     microwave
    -0.64
     reper
    -0.63
    6666
    -0.63
    POSITIVE LOGITS
     Moss
    1.10
    es
    0.86
    berg
    0.81
    creen
    0.80
    olini
    0.76
    ett
    0.73
     Lerner
    0.70
    omon
    0.68
     Rocks
    0.68
    endon
    0.68
    Activations Density 0.009%

    No Known Activations