INDEX

Explanations

phrases that indicate a basis for decision-making or judgment

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

-0.57

WriteBarrier

-0.55

 Extragalactic

-0.54

 manières

-0.53

JvmStatic

-0.52

 anlam

-0.52

 inneces

-0.51

 Lordships

-0.50

 destinées

-0.50

endphp

-0.50

POSITIVE LOGITS

 observations

0.95

 interviews

0.83

observations

0.81

 feedback

0.77

 observation

0.76

 principles

0.76

 experiences

0.72

 considerations

0.72

 conversations

0.69

 current

0.69

Activations Density 0.780%

phrases that indicate a basis for decision-making or judgment

No Comments

No Known Activations