INDEX

Explanations

names of specific locations, particularly related to Los Angeles

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Może

-0.78

Sklici

-0.75

Bardzo

-0.70

Dziękuję

-0.67

🕗

-0.64

Pozdrawiam

-0.62

Dzięki

-0.62

Wię

-0.61

Dlaczego

-0.61

 piña

-0.61

POSITIVE LOGITS

 Angeles

1.20

 ANGELES

0.97

Los

0.86

LA

0.82

 California

0.81

 angeles

0.74

 Kalifor

0.73

California

0.72

LA

0.72

Los

0.71

Activations Density 0.062%

names of specific locations, particularly related to Los Angeles

No Comments

No Known Activations