© 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. 16414
    Prev
    Next
    INDEX
    Explanations

    terms related to being oriented towards a specific target or goal

    oai_token-act-pair · gpt-3.5-turbo
    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
     £
    -0.71
     Cum
    -0.70
     nails
    -0.67
     Pompe
    -0.65
     cens
    -0.64
     donation
    -0.61
     tum
    -0.61
     Getty
    -0.60
     Britann
    -0.59
     Donation
    -0.58
    POSITIVE LOGITS
    oriented
    3.59
    orient
    2.50
     oriented
    2.18
    focused
    1.98
    centered
    1.62
    centric
    1.49
     orient
    1.43
    driven
    1.32
    organized
    1.29
    minded
    1.28
    Activations Density 0.013%

    No Known Activations