© 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. 8059
    Prev
    Next
    INDEX
    Explanations

    references to metal objects

    oai_token-act-pair · gpt-3.5-turbo

    occurrences of the word "metal" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    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
    isSpecialOrderable
    -0.86
    utt
    -0.78
    afort
    -0.78
    Plugin
    -0.72
    VPN
    -0.72
    soDeliveryDate
    -0.71
    opp
    -0.70
    otiation
    -0.68
    APTER
    -0.66
     Fidel
    -0.66
    POSITIVE LOGITS
     metal
    3.59
    metal
    2.81
    Metal
    2.50
     metals
    2.48
     Metal
    2.46
     metallic
    2.10
     steel
    1.72
     aluminum
    1.68
     aluminium
    1.66
     titanium
    1.52
    Activations Density 0.019%

    No Known Activations