© 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. 23986
    Prev
    Next
    INDEX
    Explanations

    words associated with identification or categorization, such as markers or gender markers

    oai_token-act-pair · gpt-3.5-turbo

    references to markers that signify important or distinguishing features in various contexts

    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
    erest
    -0.86
    orld
    -0.84
    ibaba
    -0.78
    obbies
    -0.77
    é¾
    -0.77
    ILLE
    -0.76
    ategory
    -0.75
    rina
    -0.74
    acia
    -0.74
    awar
    -0.74
    POSITIVE LOGITS
     marker
    1.34
     markers
    1.24
    posts
    0.85
     marking
    0.79
    holder
    0.76
     plaque
    0.72
     dotted
    0.69
     pens
    0.69
     flare
    0.69
     indicating
    0.68
    Activations Density 0.007%

    No Known Activations