INDEX

Explanations

references to expertise and specialized skills within various professional contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

期刊论文

-0.50

 colis

-0.45

TouchListener

-0.41

ClientRect

-0.39

raborty

-0.37

 TextAlign

-0.37

TouchEvent

-0.36

 Predict

-0.36

Frequency

-0.36

colli

-0.36

POSITIVE LOGITS

 expertise

0.72

 specialists

0.68

 specialist

0.66

 skills

0.65

 especialista

0.64

expertise

0.58

specialist

0.58

 EXPERT

0.58

Expertise

0.57

 expert

0.57

Activations Density 0.031%

references to expertise and specialized skills within various professional contexts

No Comments

No Known Activations