INDEX

Explanations

phrases indicating recommendations or advice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

XmlAccessorType

-0.62

 Barrera

-0.61

iczna

-0.57

 препратки

-0.57

igay

-0.57

TextAlign

-0.55

 Abbas

-0.53

inola

-0.52

 turg

-0.52

textAlignment

-0.52

POSITIVE LOGITS

 préférable

0.93

ควร

0.90

 should

0.88

 advisable

0.87

should

0.85

 faudrait

0.85

 sebaiknya

0.85

 conviene

0.82

 Should

0.80

 следует

0.77

Activations Density 0.305%

phrases indicating recommendations or advice

No Comments

No Known Activations