INDEX

Explanations

the concept of sameness or similarity in various contexts

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

 apalagi

-0.56

 الحره

-0.56

TRIBUN

-0.52

 actualidad

-0.51

Similarly

-0.47

seamnă

-0.45

 préfé

-0.45

espère

-0.45

 "..\..\..\

-0.45

 régulièrement

-0.45

POSITIVE LOGITS

 exact

1.23

exact

0.97

 Exact

0.97

 EXACT

0.92

 thing

0.87

 kind

0.85

 amount

0.80

Exact

0.80

EXACT

0.79

 applies

0.74

Activations Density 0.117%

the concept of sameness or similarity in various contexts

No Comments

No Known Activations