INDEX

Explanations

various forms of the word "str"

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

Geplaatst

-0.97

 TextInputType

-0.71

 برانيه

-0.69

CloseOperation

-0.68

Obrázky

-0.67

styleType

-0.66

 HasFactory

-0.65

pensive

-0.65

ghest

-0.64

Попис

-0.63

POSITIVE LOGITS

STR

0.61

 strs

0.60

 linkovi

0.54

Str

0.52

str

0.52

 Stroud

0.47

 Kens

0.47

inflate

0.46

str

0.45

cases

0.45

Activations Density 0.096%

various forms of the word "str"

No Comments

No Known Activations