INDEX

Explanations

terms related to prompts and reactions in programming contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 2-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

Personensuche

-0.48

 resourceCulture

-0.44

 kasarigan

-0.43

حياته

-0.42

XXV

-0.42

 AssemblyVersion

-0.42

tun

-0.40

 willy

-0.39

 Berk

-0.39

stre

-0.39

POSITIVE LOGITS

 React

0.91

 react

0.83

 prompt

0.80

 Prompt

0.79

prompt

0.75

React

0.73

Prompt

0.73

react

0.65

 prompts

0.65

 trait

0.64

Activations Density 0.098%

terms related to prompts and reactions in programming contexts

No Comments

No Known Activations