INDEX

Explanations

phrases indicating a lack of progress or realization regarding various concepts or technologies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 ſeveral

-0.89

 pleaſure

-0.79

ſelf

-0.78

 myſelf

-0.77

 purpoſe

-0.74

 itſelf

-0.73

 leſs

-0.72

 ſmall

-0.72

 miſ

-0.71

 ſtate

-0.71

POSITIVE LOGITS

 adequately

0.95

 vraiment

0.83

 véritable

0.82

 really

0.81

Yet

0.79

yet

0.78

 sufficiently

0.76

any

0.76

 seem

0.75

 adequate

0.72

Activations Density 0.673%

phrases indicating a lack of progress or realization regarding various concepts or technologies

No Comments

No Known Activations