INDEX

Explanations

references to studies or documents with specific formatting cues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

 Mikey

-0.96

 Eilish

-0.90

farwyddwr

-0.87

 متعلقه

-0.83

存于互联网档案馆

-0.82

Alam

-0.79

 resourceCulture

-0.78

WindowConstants

-0.78

Mikey

-0.78

getch

-0.77

POSITIVE LOGITS

Convex

0.89

convex

0.88

 convex

0.87

Hull

0.86

 Hull

0.80

hull

0.77

 Convex

0.76

 lait

0.74

 hull

0.71

 Kirk

0.71

Activations Density 0.088%

references to studies or documents with specific formatting cues

No Comments

No Known Activations