INDEX

Explanations

references to specific timing or locations in narratives

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

ورية

-0.53

arel

-0.53

ssysteme

-0.53

adpleegd

-0.51

ClassNotFound

-0.48

zuführen

-0.46

inverno

-0.46

 Alegre

-0.45

 resourceCulture

-0.45

 dàng

-0.44

POSITIVE LOGITS

 nakalista

0.75

 BoxFit

0.74

MLLoader

0.66

fjspx

0.65

 utafitiHapana

0.62

atguigu

0.61

 InputDecoration

0.58

WriteTagHelper

0.56

BagConstraints

0.56

 ComVisible

0.55

Activations Density 0.139%

references to specific timing or locations in narratives

No Comments

No Known Activations