INDEX

Explanations

code structure and function definitions

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

 mole

-0.54

 pymongo

-0.48

 khảo

-0.47

 Surv

-0.45

 Mole

-0.44

ول

-0.43

ỡng

-0.43

 committee

-0.42

melidir

-0.42

梅

-0.42

POSITIVE LOGITS

 registers

0.79

 instruction

0.78

 Instruction

0.76

 address

0.75

WriteBarrier

0.72

 Италијани

0.72

 disassembly

0.71

ContentAlignment

0.71

 Address

0.71

 Registers

0.71

Activations Density 1.437%

code structure and function definitions

No Comments

No Known Activations