INDEX

Explanations

commands and technical instructions in a document, focusing on programming related phrases

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

 philanth

-1.08

 encomp

-1.07

 depic

-1.02

 impra

-1.02

 Confu

-1.02

 volunte

-1.01

 accla

-1.01

 intersper

-1.01

 reluct

-1.00

 sophistic

-0.98

POSITIVE LOGITS

 altogether

0.97

 vůbec

0.95

 überhaupt

0.92

 вообще

0.85

 ogóle

0.78

<bos>

0.76

 even

0.68

even

0.66

 siquiera

0.65

 כלל

0.60

Activations Density 0.338%

commands and technical instructions in a document, focusing on programming related phrases

No Comments

No Known Activations