INDEX

Explanations

terms related to ceilings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

Efq

-0.70

espère

-0.63

 isolado

-0.63

 entanto

-0.61

 Ahnung

-0.60

 honte

-0.58

NOPQRST

-0.57

 Monter

-0.57

 McClure

-0.55

 Encyclopædia

-0.55

POSITIVE LOGITS

 ceiling

0.73

RenderAtEndOf

0.72

Ron

0.72

 ceilings

0.71

subset

0.70

 wanna

0.65

 Ceiling

0.65

addr

0.64

subseteq

0.64

ceiling

0.64

Activations Density 0.120%

terms related to ceilings

No Comments

No Known Activations