INDEX

Explanations

numeric identifiers and related information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

ítulo

-0.56

 later

-0.53

<eos>

-0.52

celli

-0.49

Dal

-0.49

 gepubliceerd

-0.49

 برانيه

-0.48

كتشاف

-0.45

érables

-0.45

 Later

-0.45

POSITIVE LOGITS

Efq

0.81

 <>",

0.74

 Shakspeare

0.70

 createState

0.67

awtextra

0.66

 pleaſure

0.65

 Савезне

0.65

 Houſe

0.64

 Majefty

0.63

 myſelf

0.63

Activations Density 0.091%

numeric identifiers and related information

No Comments

No Known Activations