INDEX

Explanations

terms related to null values and boolean states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

+#+#

-0.87

 виправивши

-0.80

########.

-0.77

uxxxx

-0.69

 ujednoznacz

-0.69

+:+

-0.69



-0.69

تقاوى

-0.68

 snippetHide

-0.68

 مرئيه

-0.67

POSITIVE LOGITS

NonNull

0.42

 mandatory

0.42

NotNull

0.39

Pen

0.37

Pen

0.36

 compulsory

0.36

fal

0.35

 double

0.35

 strong

0.35

IsRequired

0.35

Activations Density 0.000%

terms related to null values and boolean states

No Comments

No Known Activations