INDEX

Explanations

occurrences of specific symbols, particularly "<bos>"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_65k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.12.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

ISupport

-0.67

aarrggbb

-0.59

iek

-0.55

zu

-0.53

eder

-0.52

ple

-0.52

ApiModelProperty

-0.52

hote

-0.52

לה

-0.52

mb

-0.50

POSITIVE LOGITS

anmoins

0.73

zędu

0.59

 useAppContext

0.57

 Pourtant

0.57

かわらず

0.56

\{\\

0.55

 enfans

0.52

 stället

0.52

ępnie

0.51

 Toutefois

0.50

Activations Density 0.230%

occurrences of specific symbols, particularly "<bos>"

No Comments

No Known Activations