INDEX

Explanations

phrases related to making requests or seeking help

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.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

 richting

-0.45

 degenerate

-0.41

complexContent

-0.41

 princípios

-0.38

setFlags

-0.38

ComponentScan

-0.38

 Partially

-0.37

 ubicación

-0.37

 voorbere

-0.37

 orient

-0.37

POSITIVE LOGITS

 requesting

0.74

 request

0.73

 requested

0.71

 nakalista

0.70

 Request

0.67

request

0.66

requested

0.65

 REQUEST

0.63

 Requested

0.63

Requested

0.62

Activations Density 0.421%

phrases related to making requests or seeking help

No Comments

No Known Activations