INDEX

Explanations

quantitative data and statistics related to performance

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

CloseOperation

-0.94

 AttributeSet

-0.86

 HasFactory

-0.84

 AssemblyCulture

-0.78

 <=",

-0.74

 Phry

-0.73

 Dalai

-0.72

enumi

-0.72

 ddelweddau

-0.71

 Quelles

-0.69

POSITIVE LOGITS

 popolari

0.80

 popular

0.67

“

0.63

 egne

0.63

 flagship

0.63

 quod

0.61

هما

0.60

による

0.60

しまって

0.60

popular

0.59

Activations Density 0.238%

quantitative data and statistics related to performance

No Comments

No Known Activations