INDEX

Explanations

words related to praise or positive evaluation

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

 Minang

-0.96

 seksi

-0.90

 praktik

-0.86

 panik

-0.85

 maksi

-0.85

 Batam

-0.82

 keramik

-0.82

 kram

-0.81

 Karang

-0.81

 simplif

-0.81

POSITIVE LOGITS

as

0.78

well

0.64

 WELL

0.63

as

0.61

 well

0.60

Well

0.59

 Well

0.56

WarningLevel

0.52

WELL

0.52

 wells

0.52

Activations Density 0.081%

words related to praise or positive evaluation

No Comments

No Known Activations