INDEX

Explanations

phrases related to falsehoods or inaccuracies

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

HttpPut

-0.48

 Muffins

-0.46

<0xDF>

-0.46

 للاسماء

-0.45

addCriterion

-0.44

راسیون

-0.43

HPP

-0.43

IsUnicode

-0.42

HttpDelete

-0.41

Dif

-0.41

POSITIVE LOGITS

 maksi

0.87

 false

0.81

 kasa

0.78

 jati

0.78

false

0.77

 False

0.77

 Jasa

0.74

 seksi

0.72

 Palembang

0.71

 lemp

0.71

Activations Density 0.061%

phrases related to falsehoods or inaccuracies

No Comments

No Known Activations