INDEX

Explanations

details related to precision and accuracy in information

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

 affor

-0.91

 increa

-0.90

 volunte

-0.89

 intersper

-0.85

 inev

-0.85

 Intere

-0.84

 impra

-0.84

 unwarran

-0.84

fuf

-0.83

 emphat

-0.82

POSITIVE LOGITS

 accuracy

1.25

 accurate

1.20

 ACCUR

1.16

accuracy

1.09

 Accuracy

1.09

Accuracy

1.06

accurate

1.05

 Accurate

1.04

 inaccurate

0.97

Accurate

0.96

Activations Density 0.080%

details related to precision and accuracy in information

No Comments

No Known Activations