INDEX

Explanations

sections formatted with headers or notable markers like bold and italicized text

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_2/width_16k/average_l0_141

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

InputTagHelper

-0.76

хьтан

-0.75

 kasarigan

-0.73

SequentialGroup

-0.70

 Roskov

-0.70

EndGlobalSection

-0.67

contentLoaded

-0.66

MessageTagHelper

-0.65

Referanser

-0.64

 referenties

-0.62

POSITIVE LOGITS

posedge

0.57

uyez

0.53

idän

0.50

первых

0.50

WriteTagHelper

0.48

FRONT

0.47

 voors

0.47

toHaveBeen

0.47

Londres

0.47

pexpr

0.46

Activations Density 0.291%

sections formatted with headers or notable markers like bold and italicized text

No Comments

No Known Activations