INDEX

Explanations

statements that indicate whether something is logical or understandable

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

 gesteld

-0.66

tagHelper

-0.63

 Albion

-0.63

 betreft

-0.60

 ospiti

-0.57

cèse

-0.56

={`/

-0.55

 lifestyles

-0.54

 بث

-0.54

 occasions

-0.53

POSITIVE LOGITS

 nonsense

0.81

nonsense

0.81

Nonsense

0.80

ensical

0.76

coher

0.67

 logique

0.65

onsense

0.64

 gibber

0.63

 logical

0.62

 sensible

0.60

Activations Density 0.171%

statements that indicate whether something is logical or understandable

No Comments

No Known Activations