INDEX

Explanations

references to specific numerical counts and positions in a sequence

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

 liev

-1.00

 strass

-0.94

 anse

-0.93

 naer

-0.87

 adal

-0.87

 utop

-0.85

 maksi

-0.85

 ordina

-0.83

 saar

-0.83

ymm

-0.83

POSITIVE LOGITS

 ones

0.80

 Incluso

0.60

 Ceux

0.56

 recent

0.54

 unspeak

0.54

 Algunas

0.54

 recently

0.53

 including

0.53

 repugnant

0.52

 Aún

0.52

Activations Density 0.393%

references to specific numerical counts and positions in a sequence

No Comments

No Known Activations