INDEX

Explanations

references to the concept of discovery or findings in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 отношению

-0.73

-0.69

-0.68

ktm

-0.66

-0.65

-0.63

خدام

-0.63

-0.62

-0.61

POSITIVE LOGITS

 discovery

2.05

 discoveries

2.00

 Discovery

1.84

 Discover

1.80

Discovery

1.72

DISCOVER

1.68

discover

1.64

discovery

1.64

 discover

1.63

 discovers

1.62

Activations Density 0.044%

references to the concept of discovery or findings in various contexts

No Comments

No Known Activations