INDEX

Explanations

terms related to machine learning and chemical processes

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 myſelf

-1.20

 themſelves

-1.09

 himſelf

-1.08

 juſt

-0.97

 pleaſure

-0.96

 Theſe

-0.95

Efq

-0.94

 itſelf

-0.93

 hâte

-0.93

 Footh

-0.93

POSITIVE LOGITS

 machines

0.77

<b>

0.73

化

0.73

Machines

0.69

<h2>

0.68

work

0.67

 machine

0.67

<i>

0.66

0.63

0.62

Activations Density 0.096%

terms related to machine learning and chemical processes

No Comments

No Known Activations