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. 3-RES-JB
    6. 13971
    Prev
    Next
    INDEX
    Explanations

    related to instances of causality or things that come before other things

    oai_token-act-pair · gpt-3.5-turbo

    terms related to causation and predictive relationships

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GPT2-SMALL @ 3-res-jb
    Configuration
    jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    Skylion007/openwebtext
    Hook Point Layer
    3
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    aneers
    -0.91
    arro
    -0.79
    quished
    -0.79
    enstein
    -0.77
    atan
    -0.77
    estern
    -0.77
    tower
    -0.76
    oise
    -0.76
    ÄŁ
    -0.76
    ffen
    -0.75
    POSITIVE LOGITS
     predictive
    0.91
     precursor
    0.85
     indicative
    0.84
     indicators
    0.83
    interstitial
    0.79
     satell
    0.78
    hetically
    0.78
     opio
    0.76
     correlate
    0.76
     markers
    0.75
    Activations Density 0.024%

    No Known Activations