INDEX

Explanations

medical terms related to cancer and treatments in text related to technology and community recommendations

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

Dział

-0.74

Примеча

-0.73

Skład

-0.73

Przyp

-0.70

tidaknya

-0.70

Witaj

-0.69

Vezi

-0.68

Ży

-0.68

 Divulgação

-0.68

 piña

-0.68

POSITIVE LOGITS

etc

0.89

etc

0.73

usw

0.70

等等

0.61

0.60

Etc

0.59

등

0.58

0.56

stb

0.53

等

0.53

Activations Density 0.451%

medical terms related to cancer and treatments in text related to technology and community recommendations

No Comments

No Known Activations