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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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 Local
    5. 6-RES_SCL-AJT
    6. 13607
    Prev
    Next
    INDEX
    Explanations

    code syntax elements and computer-related commands

    oai_token-act-pair · gpt-3.5-turbo
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-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
    325
    +0.09
    0.5%
    123
    +0.08
    0.5%
    161
    +0.08
    0.5%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    326
    +0.09
    0.20
    288
    +0.08
    0.20
    730
    +0.08
    0.09
    Negative Logits
    rish
    -0.64
    ular
    -0.57
    pop
    -0.56
    oted
    -0.55
    ovy
    -0.55
    apore
    -0.55
    Wars
    -0.54
    tale
    -0.53
    apult
    -0.53
    coming
    -0.52
    POSITIVE LOGITS
    éĹĺ
    0.56
    ++++++++
    0.53
    -+
    0.52
     unfocusedRange
    0.52
    âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ
    0.51
    +++
    0.50
    essor
    0.48
    ensional
    0.48
    +(
    0.48
    ;;;;
    0.48
    Activations Density 6.284%

    No Known Activations