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

    Neuronpedia

    APIAssistant AxisNEWCircuit 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. 0-RES-JB
    6. 4598
    Prev
    Next
    INDEX
    Explanations

    proper nouns, phrases, or terms that are widely acknowledged or accepted

    oai_token-act-pair · gpt-3.5-turbo

    instances of the word "recognized" and its variations

    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
    tank
    -0.83
    £
    -0.82
    ageddon
    -0.74
     mum
    -0.71
    abouts
    -0.69
     £
    -0.69
    alys
    -0.69
    bs
    -0.69
     Ashes
    -0.67
    fing
    -0.65
    POSITIVE LOGITS
     recognized
    3.65
     recognised
    2.81
     recognize
    2.31
     recognizes
    2.30
    recogn
    2.18
     recognizing
    2.05
     Recogn
    1.86
     recognition
    1.84
     acknowledged
    1.75
     recognizable
    1.67
    Activations Density 0.010%

    No Known Activations