INDEX

Explanations

non-empty lists or collections in programming structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 walls

-0.63

LEncoder

-0.59

 tartalomajánló

-0.58

ymal

-0.58

-+

-0.58

tlere

-0.56

amur

-0.56

łby

-0.55

antism

-0.55

 Kort

-0.55

POSITIVE LOGITS

InjectAttribute

0.80

ftagPool

0.72

]")]

0.67

ьаж

0.65

sizeCache

0.61

UnknownFields

0.61

Demografía

0.57

 nyelven

0.55

}]);

0.55

Activations Density 0.200%

non-empty lists or collections in programming structures

No Comments

No Known Activations