INDEX

Explanations

information related to academic or scientific credentials and accomplishments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 houſe

-1.14

 ſtate

-1.03

 purpoſe

-0.96

 pleaſure

-0.94

 faſt

-0.94

 cauſe

-0.92

 leaſt

-0.91

 Houſe

-0.91

 ſmall

-0.89

 ſeveral

-0.89

POSITIVE LOGITS

rungsseite

0.70

 استنادى

0.67

__':

0.64

abestanden

0.62

 estekak

0.62

Transkript

0.61

Hochspringen

0.56

ko

0.55

__':

0.55

DBNull

0.54

Activations Density 1.138%

information related to academic or scientific credentials and accomplishments

No Comments

No Known Activations