INDEX

Explanations

keywords related to specific locations and names 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 @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Hampden

-0.94

IBOutlet

-0.89

icoot

-0.86

 McIntyre

-0.84

 toys

-0.80

djangoproject

-0.78

 Malden

-0.78

возь

-0.76

 Bachchan

-0.75

 Schilling

-0.75

POSITIVE LOGITS

verläs

0.82

 McNally

0.74

='".$

0.73

 '".$

0.72

boldsymbol

0.72

McE

0.71

Cyn

0.71

 Jubilee

0.70

 Bala

0.69

LayoutStyle

0.68

Activations Density 2.410%

keywords related to specific locations and names in various contexts

No Comments

No Known Activations