INDEX

Explanations

expressions indicating quality and thoroughness in assessments or evaluations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 Monfieur

-0.77

 Jefus

-0.72

 ſeveral

-0.72

esModule

-0.68

 aveug

-0.67

 Theſe

-0.67

 fubject

-0.66

 greateſt

-0.66

 iſt

-0.65

TagMode

-0.65

POSITIVE LOGITS

 crafted

0.93

 constructed

0.88

 designed

0.88

 timed

0.87

 organized

0.85

 worded

0.82

 structured

0.78

 arranged

0.78

 executed

0.75

 done

0.75

Activations Density 0.195%

expressions indicating quality and thoroughness in assessments or evaluations

No Comments

No Known Activations