INDEX

Explanations

statements emphasizing clarity and certainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

 Marsden

-0.73

 يتيمه

-0.72

Collegamenti

-0.68

 interchangeably

-0.66

 automatiquement

-0.65

 chivalry

-0.64

itesse

-0.63

 parlent

-0.61

 poda

-0.61

щению

-0.60

POSITIVE LOGITS

 clearly

0.96

clearly

0.92

 obvious

0.91

 evident

0.88

 Clearly

0.87

Clearly

0.87

 CLEAR

0.86

 ràng

0.81

 clear

0.80

obvious

0.77

Activations Density 0.159%

statements emphasizing clarity and certainty

No Comments

No Known Activations