INDEX

Explanations

phrases indicating conclusions or results

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

WSGI

-0.54

 groupBox

-0.52

chec

-0.51

 XMLHttpRequest

-0.50

sû

-0.49

 którzy

-0.49

říklad

-0.48

 essi

-0.47

겠

-0.47

/*:

-0.47

POSITIVE LOGITS

 being

1.00

 needing

0.95

 having

0.93

CppMethod

0.74

 wanting

0.73

 looking

0.72

being

0.71

 doing

0.70

 focusing

0.67

Portail

0.65

Activations Density 0.122%

phrases indicating conclusions or results

No Comments

No Known Activations