INDEX

Explanations

terms related to interactions and dialogue between people

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ביוגרפיה

-0.90

 Áng

-0.82

 Settembre

-0.78

 Ottobre

-0.75

 ideolog

-0.75

Debido

-0.74

 meras

-0.74

 Tomé

-0.73

 poliuret

-0.72

Glej

-0.72

POSITIVE LOGITS

 maneu

0.79

 ineffec

0.78

<bos>

0.77

 unve

0.74

 vainly

0.73

 gaily

0.70

 apprehen

0.70

 resurre

0.70

 OBSERVATIONS

0.69

 Confu

0.68

Activations Density 0.121%

terms related to interactions and dialogue between people

No Comments

No Known Activations