INDEX

Explanations

descriptive phrases pertaining to the quality or nature of experiences or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.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

 ModelExpression

-0.88

 itſelf

-0.87

Efq

-0.82

 Majefty

-0.82

 myſelf

-0.76

 Chriftian

-0.76

 himſelf

-0.74

ethene

-0.73

ruik

-0.72

sizeCache

-0.70

POSITIVE LOGITS

 easy

0.64

0.62

 hard

0.62

 very

0.60

 difficult

0.58

 time

0.57

 simple

0.50

the

0.50

 normal

0.49

 full

0.49

Activations Density 0.180%

descriptive phrases pertaining to the quality or nature of experiences or items

No Comments

No Known Activations