INDEX

Explanations

sentences that conclude or summarize points

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

 Надо

-0.98

Надо

-0.95

けっこう

-0.93

 надо

-0.90

Herzliche

-0.89

 Ditto

-0.86

 daß

-0.83

 היתה

-0.82

rungsseite

-0.82

^(@)

-0.80

POSITIVE LOGITS

Additionally

0.95

Furthermore

0.94

 Additionally

0.94

 Furthermore

0.89

 Cependant

0.88

 jedoch

0.88

Utiliz

0.87

 however

0.86

 Lastly

0.85

Lastly

0.85

Activations Density 0.226%

sentences that conclude or summarize points

No Comments

No Known Activations