INDEX

Explanations

sequences of statements expressing actions or sequences of events, often emphasizing personal involvement

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

 enfans

-0.81

 habet

-0.80

 auroit

-0.79

 démocr

-0.74

 avoient

-0.73

 sanitaires

-0.73

 étoit

-0.72

 domestiques

-0.70

 automatiques

-0.70

 pouvoit

-0.70

POSITIVE LOGITS

estacks

0.71

 simply

0.70

re

0.65

 quietly

0.64

PreferredItem

0.64

UAWEI

0.64

oa̍t

0.61

 AssemblyTitle

0.61

 systematically

0.60

 carefully

0.58

Activations Density 0.689%

sequences of statements expressing actions or sequences of events, often emphasizing personal involvement

No Comments

No Known Activations