INDEX

Explanations

instances of evidence or statements related to analysis and conclusions drawn from data or observations

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 pleaſure

-1.03

 Monfieur

-0.92

 houſe

-0.86

 Jefus

-0.85

 Cæsar

-0.83

 theſe

-0.81

 ſch

-0.79

 ſtate

-0.79

 ſhe

-0.77

 greateſt

-0.77

POSITIVE LOGITS

was

0.81

did

0.79

 didn

0.78

 volna

0.75

 ayer

0.75

 damals

0.74

 originally

0.73

 previously

0.73

 wasn

0.72

 came

0.72

Activations Density 6.137%

instances of evidence or statements related to analysis and conclusions drawn from data or observations

No Comments

No Known Activations