Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Joseph Bloom · Open Source Sparse Autoencoders for all Residual Stream Layers of GPT2-Small
    3. GPT2-Small
    4. Residual Stream
    5. 4-RES-JB
    6. 5074
    Prev
    Next
    INDEX
    Explanations

    and focus on strengths and weaknesses

    oai_token-act-pair · gpt-3.5-turbo

    sections that evaluate strengths and weaknesses in a competitive context

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

    No Comments

    Negative Logits
    ĸļ
    -0.84
     tremend
    -0.76
    ozo
    -0.75
     inward
    -0.75
    oud
    -0.74
    yout
    -0.73
    undai
    -0.73
    rency
    -0.73
    bered
    -0.73
    ulz
    -0.72
    POSITIVE LOGITS
     TBD
    0.98
    ³³³³³³³³³³³³³³³³
    0.91
     Scores
    0.85
     Provided
    0.81
     Measures
    0.78
     Where
    0.75
     Named
    0.75
     Runs
    0.75
     Starts
    0.74
     Unknown
    0.73
    Activations Density 0.153%

    No Known Activations