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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Apollo Research · Taylor · Identifying Functionally Important Features with End-to-End Sparse Dictionary Learning
    3. GPT2-Small
    4. Residual Similar CE E2E Recon
    5. 6-RES_SCEFR-AJT
    6. 1017
    Prev
    Next
    INDEX
    Explanations

    legal and regulatory terms tied to various industries and contexts

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-ajt
    Prompts (Dashboard)
    12,288 prompts, 128 tokens each
    Dataset (Dashboard)
    Skylion007/openwebtext
    Features
    46,080
    Data Type
    torch.float32
    Hook Point
    blocks.6.hook_resid_pre
    Architecture
    standard
    Context Size
    128
    Dataset
    apollo-research/Skylion007-openwebtext-tokenizer-gpt2
    Hook Point Layer
    6
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    160
    +0.10
    0.6%
    290
    +0.10
    0.6%
    152
    +0.09
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    152
    +0.10
    0.13
    124
    +0.10
    0.10
    537
    +0.09
    0.10
    Negative Logits
     anecd
    -0.55
     Gillespie
    -0.55
     adversely
    -0.53
     privately
    -0.50
     beads
    -0.50
     incess
    -0.49
    kson
    -0.49
     Dickinson
    -0.49
     furiously
    -0.49
     dissenting
    -0.48
    POSITIVE LOGITS
    level
    0.84
    life
    0.81
    type
    0.78
    motion
    0.77
    feeding
    0.76
    centered
    0.74
    based
    0.74
    prototype
    0.74
    through
    0.74
    effects
    0.73
    Activations Density 0.573%

    No Known Activations