INDEX

Explanations

phrases that refer to comparisons indicating a lower quantity or degree

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

LEGGI

-0.76

 Pingback

-0.72

 doInBackground

-0.71

mentaires

-0.70

Personensuche

-0.70

 BoxDecoration

-0.67

 iſt

-0.67

felves

-0.66

Portale

-0.66

})`

-0.65

POSITIVE LOGITS

 amount

0.71

 than

0.69

ens

0.66

(<

0.62

 less

0.61

 LESS

0.61

ening

0.59

 Less

0.57

#+#

0.56

than

0.55

Activations Density 0.093%

phrases that refer to comparisons indicating a lower quantity or degree

No Comments

No Known Activations