INDEX

Explanations

phrases related to negativity, criticism, or unfortunate situations

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

Související

-0.67

Voci

-0.62

municipi

-0.60

Trotz

-0.54

Sunt

-0.53

dirent

-0.52

Läs

-0.52

hatenablog

-0.52

cnbc

-0.51

Agrade

-0.51

POSITIVE LOGITS

 malang

0.65

 unfortunately

0.65

 wretch

0.63

 Mère

0.63

 Unfortunately

0.60

 Traps

0.59

unfortunately

0.59

 nmax

0.59

heureusement

0.59

 Stretcher

0.57

Activations Density 0.098%

phrases related to negativity, criticism, or unfortunate situations

No Comments

No Known Activations